managing water as a constraint to development with ... · of cape town (uct) african climate and...
TRANSCRIPT
Managing Water as a Constraint to Development with Decision-Support
Tools That Promote Integrated Planning: The Case of the Berg
Water Management Area
C. Pengelly, H. Seyler, N. Fordyce, P. Janse van Vuuren, M. van der
Walt, H. van Zyl & J. Kinghorn
Report to the
Water Research Commission and Western Cape Government
by
GreenCape Sector Development Agency
WRC Report No. TT xxx/xx
December 2017
ii
Obtainable from
Water Research Commission
Private Bag X03
Gezina, 0031
South Africa
The publication of this report emanates from a project entitled Towards Sustainable Economic
Development in Water Constrained Catchments: Tools to Empower Decision Making (WRC Project No.
K5/2453)
ISBN XXXXX
Printed in the Republic of South Africa
DISCLAIMER
This report has been reviewed by the Water Research Commission (WRC) and approved for
publication. Approval does not signify that the contents necessarily reflect the views and policies of
the WRC, nor does mention of trade names or commercial products constitute endorsement or
recommendation for use.
iii
Executive summary There is increasing recognition that the combined effects of climate change, population growth and
continued urbanisation are exerting pressure on limited water resources. At the same time, economic
growth remains vital for alleviating poverty. Therefore, growth is required in spite of significant water
resource constraints. At issue then is how to allocate water optimally to enable economic growth, while
also ensuring that human needs are met and ecological systems maintained.
Understanding of the economic impacts of water access is limited, and there is a lack of tools available
to address the trade-offs that may be required when allocating water in a water scarce system. This is
needed in particular in “constrained catchments” where all readily available water is already allocated,
such as the case of the Berg Water Management Area (WMA). In such catchments, future development
requires additional water resources, either through the development of new resources or the
reallocation from other users in the WMA.
This three-year study (Managing Water as a Constraint to Development with Decision-Support Tools
That Promote Integrated Planning: The Case of the Berg Water Management Area) aimed to better
integrate water into economic development planning, and vice versa. The project’s objectives included:
developing a guideline for a planning approach that recognised the cyclical interdependency of
economics and water resources; conducting a cost-benefit analysis of economic developments and
water resource interventions; building a spatial hydro-economic tool to manage regional allocations in
constrained catchments; and developing research products in close collaboration with decision-makers;
and implementing research outcomes to address current development challenges. These objectives
have been achieved through developing actionable insights and tools for governmental decision-
makers that link water as a resource and its availability to economic outcomes in terms of growth and
job creation. The project structure was highly collaborative, lead by a sector development agency for
the green economy (GreenCape) and utilising research generated by Masters students at the University
of Cape Town (UCT) African Climate and Development Institute (ACDI) and adapting the approach and
outputs to the evolving requirements of government stakeholders. The Berg WMA was selected as a
case study area.
Analysis was done on the (potential for) integration between development planning, water allocation
and water resource development processes, with an emphasis on enabling implementation of the
findings. The analysis aimed to understand how decision support tools could add value to existing
legislated processes by filling knowledge gaps or by providing a collaboration mechanism. This analysis
highlighted how the Integrated Development Plan (IDP) is the key integrated planning tool, but that it is
not taking water resource availability sufficiently into account. Furthermore, that the municipalities,
which are also the Water Service Authorities (WSAs), do not have the capacity nor resources to develop
their own local water resources, and are struggling to access water from the regional schemes,
managed nationally by the Department of Water and Sanitation (DWS). In this context, provincial
government has a critical coordinating and supporting role to play.
A regional hydro-economic GIS tool has been developed to understand how water scarcity may
constrain development in local economies within the Berg WMA, currently and into the future, including
due exogenous factors such as climate change and population growth. The study models future water
demand, focusing on the years 2025 and 2040 to assess where demand may outstrip supply according
to the current system yield. These projected demands link water usage to economic indicators (e.g.
gross value add (GVA), jobs) to highlight where these constraints have the most significant economic
implications, thereby allowing for the prioritisation of interventions to improve water supply to particular
local economies. The results indicate that under all climate change models, irrigated agriculture will
require more water to remain sustainable. However, at the same time, urban centres will demand
iv
increasingly more water. The water required for urban areas generates greater value to the regional
economy, with the City of Cape Town continuing to dominate regional economic outcomes. However,
when water is analysed as a constraint to the local economy, the West Coast municipalities of
Swartland, Saldanha Bay and Bergrivier emerge as the municipalities where water is likely be to a
significant constraint to future development (with substantial impact on the local economy and
livelihoods), unless water supply augmentation options are developed urgently. However, arguably, the
current drought in Western Cape has already illustrated the economic impacts of reduced water
availability, as well as the need for diversified local water supply that is not solely reliant on surface
water.
In the case of Saldanha Bay, where water is already a constraint to development, a Multi-Criteria
Decision Analysis (MCDA) tool was developed that allows the municipality to prioritise new development
applications on the basis of the socio-economic outcomes from the various projects in comparison to
the water required, rather than just allocate water on a first-come-first-served basis. This tool allows for
transparent and collaborative decision-making that assists the municipality in improving the livelihoods
of the local community and increasing the productivity of water in the local economy.
The project outcomes have been shared with local stakeholders responsible for development planning
and water management. The tools have been favourably received with provincial government
departments declaring their intent to adopt them in 2018. The tools are also easily replicable for other
regions in South Africa that face similar challenges and wish to better integrate their water resource and
development planning. The project also highlighted gaps in the availability of data to be able to develop
a fully-functional hydro-economic tool. These gaps include a sectoral breakdown of water usage in
urban areas, multi-year agricultural revenue and production figures, and metered agricultural water
usage.
Overall the project has been beneficial in helping drive a dialogue amongst a number of stakeholders
of the importance of water for development. It has elevated the consideration of water within provincial
government from an environmental concern to one that enables growth and jobs. It has also helped
substantiate the need for urgent intervention in certain areas for supply augmentation.
However, the project findings highlight the need for further action:
The absence of a regional water utility for the Berg WMA is hampering the development of
water resources, particularly at a local level, and the creation of either a regional water utility or
water board should be explored in an intergovernmental process that results in an
implementation protocol. An implementation protocol will allow local, provincial and national
government departments to practically coordinate the work required to set up a suitable regional
entity.
Coordinated planning for future water resource interventions for those WSAs supplied by the
Western Cape Water Supply System (WCWSS) must be strengthened. A feasibility study is
proposed to estimate the economies of scale and efficiencies that could be gained in combining
regional and local water resource development schemes across the Berg WMA.
A regional (or national) fund that WSAs can access for off-budget water resource infrastructure,
which leverages private sector funds, should be explored
All spheres of government need to be more cognisant of the local capabilities and water
resource availability in the areas in which development is planned, crucially as it relates to the
availability of water to support their development ambitions. If sufficient water is not available,
then a careful consideration should be made about the suitability of a particular industry for the
area (linked to its water intensity) versus the cost/benefit of developing new water resources.
v
Contents Executive summary .................................................................................................................................iii
List of figures .......................................................................................................................................... vi
List of tables ........................................................................................................................................... ix
Acknowledgements ................................................................................................................................. x
List of acronyms ..................................................................................................................................... xi
1. Introduction ...................................................................................................................................... 1
1.1. Study Vision .............................................................................................................................. 1
1.2. Study Motivation ....................................................................................................................... 1
1.3. Previous/Related Work ............................................................................................................. 2
1.4. Focus on Berg Water Management Area and Saldanha Bay .................................................. 3
1.5. Research aims and proposed tools .......................................................................................... 4
1.6. Study spatial boundaries and focus areas ............................................................................... 4
1.7. Outline of this document ........................................................................................................... 6
2. Methodology .................................................................................................................................... 7
2.1. Project approach ....................................................................................................................... 7
2.2. Co-production with decision-makers ........................................................................................ 7
3. Study Area Background and Challenges ....................................................................................... 10
3.1. Water resources in the Berg Water Management Area ......................................................... 10
3.2. Water resources in Saldanha Bay Local Municipality ............................................................ 13
3.3. User Needs summary ............................................................................................................. 15
4. Status quo of water resources and economic development planning ........................................... 18
4.1. Introduction ............................................................................................................................. 18
4.2. Perspectives and practices ..................................................................................................... 18
4.3. Legislated planning processes ............................................................................................... 20
4.3.1. Water allocation ............................................................................................................... 20
4.3.2. Development planning process ....................................................................................... 27
4.3.3. Water Services and Resources Development Process .................................................. 33
4.4. Key forums for integrated water resources and economic development ............................... 40
4.4.1. Overview .......................................................................................................................... 40
4.4.2. Western Cape Sustainable Water Management Plan .................................................... 45
4.5. Insights ................................................................................................................................... 46
4.6. Way forward ............................................................................................................................ 48
5. Value of water to the regional and municipal economies .............................................................. 50
5.1. Introduction ............................................................................................................................. 50
5.2. Sectoral value of water: agricultural sector case study .......................................................... 51
5.2.1. Marginal and average values .......................................................................................... 52
5.2.2. Social value of water: employment in agriculture ............................................................ 56
5.3. Water intensity of regional and municipal economies ............................................................ 59
5.3.1. Proposed approach to understanding value of water...................................................... 59
5.3.2. Application to Berg WMA ................................................................................................ 59
5.3.3. Economic value of water at a municipal level ................................................................. 63
5.3.4. Social value of water at a municipal level ....................................................................... 64
5.4. Conclusions ............................................................................................................................ 65
6. Regional hydro-economic GIS model ............................................................................................ 67
6.1. Introduction ............................................................................................................................. 67
6.2. Estimating irrigated agriculture water requirements ............................................................... 68
6.2.1. Current irrigated water requirements .............................................................................. 69
6.2.2. Future water requirements .............................................................................................. 79
6.3. Assessment of urban water requirements .............................................................................. 89
vi
6.3.1. Current urban water requirements .................................................................................. 90
6.3.2. Future urban requirements .............................................................................................. 94
6.4. Total water requirements ........................................................................................................ 99
6.4.1. Current total water requirements ..................................................................................... 99
6.4.2. Future total water requirements .................................................................................... 101
6.5. Comparative assessment of future water requirements to current economy ....................... 103
6.5.1. Methodology .................................................................................................................. 103
6.5.2. Results ........................................................................................................................... 105
7. Local water and development allocation prioritisation ................................................................. 111
7.1. Introduction ........................................................................................................................... 111
7.2. Multi-Criteria Decision Analysis (MCDA) and municipal decision-making ........................... 111
7.3. Overview of the tool .............................................................................................................. 112
7.3.1. Selection of alternatives ................................................................................................ 113
7.3.2. Criteria selection and metrics ........................................................................................ 113
7.3.3. Criteria scoring .............................................................................................................. 119
7.3.4. Criteria weighting ........................................................................................................... 122
7.3.5. Generated results .......................................................................................................... 126
7.4. Application of tool in Saldanha Bay ...................................................................................... 127
7.4.1. Alternatives .................................................................................................................... 127
7.4.2. Criteria selection and metrics ........................................................................................ 128
7.4.3. Criteria scoring .............................................................................................................. 131
7.4.4. Criteria Weighting .......................................................................................................... 137
7.4.5. Generated Results ........................................................................................................ 139
7.5. Conclusions and insights ...................................................................................................... 141
7.6. Way forward .......................................................................................................................... 141
8. Project insights and successes .................................................................................................... 143
8.1. Analytical insights ................................................................................................................. 143
8.2. Project successes ................................................................................................................. 145
8.3. Methodology and process insights ....................................................................................... 146
8.4. Concluding Remarks ............................................................................................................ 147
9. Recommendations ....................................................................................................................... 148
9.1. Planning approaches ............................................................................................................ 148
9.2. Use of tools developed ......................................................................................................... 149
9.3. Further development of tools ................................................................................................ 150
9.4. Remaining gaps and suggestions for future research .......................................................... 150
10. References ............................................................................................................................... 151
List of figures Figure 1: Berg Water Management Area including municipal boundaries ............................................. 5
Figure 2: Project structure ....................................................................................................................... 7
Figure 3: Framework for the iterative, co-development of solutions with decision-makers .................... 9
Figure 4: Western Cape Water Supply System network infrastructure ................................................ 11
Figure 5: WCWSS reconciliation of supply and demand for the Planning Scenario (Redrawn from
DWS, 2015a) ................................................................................................................................. 12
Figure 6: Abstraction versus allocation to the Withoogte Scheme from the WCWSS (DWS, 2015a) .. 13
Figure 7: Projected SBLM water allocation and abstraction (DED&T, 2016; DWS, 2015a) ................. 14
Figure 8: Stakeholders prioritising identified challenges (shown on white posters) through voting
(coloured numbered notes) in June 2015 ...................................................................................... 17
Figure 9: Stakeholder perspectives on the relative importance of challenges identified ...................... 17
vii
Figure 10: A flow chart of the water allocation process based on analysis of legislation and verification
by stakeholders .............................................................................................................................. 22
Figure 11: A flow chart of the development planning process based on assessment of legislation and
discussion with stakeholders ......................................................................................................... 30
Figure 12: Structure of key plans working within government (adapted from DRDLR, 2012) .............. 31
Figure 13: A flow chart of the water resources and services development process based on
assessment of legislation and discussion with stakeholders ......................................................... 39
Figure 14: Image of prioritised objectives (under “cooperative governance and planning”) as voted for
by stakeholders during the SWMP mandate review workshop (May 2017) .................................. 46
Figure 15: Regional hydro-economic model components .................................................................... 51
Figure 16: Average value of irrigated agriculture per year by municipality in the Berg WMA (R millions)
....................................................................................................................................................... 55
Figure 17: Value from irrigated agriculture in Berg Water Management Area ...................................... 55
Figure 18: Gross value added and employment by water intensity of sectors in the municipalities in
the Berg Water Management Area ................................................................................................ 60
Figure 19: Gross value added and employment by water intensity of sectors in the municipalities in
Berg WMA ...................................................................................................................................... 62
Figure 19: Screenshot from SAPWAT4 (WRC, 2016) of wine grape irrigation estimate ...................... 71
Figure 20: Overview of irrigated water requirements GIS model .......................................................... 73
Figure 21: Screenshot of shapefile outputs for irrigated fields .............................................................. 74
Figure 22: Irrigated agriculture water requirements per year in the Berg WMA ................................... 75
Figure 23: Irrigated water requirements per municipality per year ....................................................... 76
Figure 24: Average water requirement per hectare per year ................................................................ 77
Figure 25: Grapes water intensity per hectare per municipality............................................................ 78
Figure 26: Mean monthly temperatures in the Berg WMA .................................................................... 80
Figure 27: Mean monthly effective rainfall in the Berg WMA ................................................................ 81
Figure 28: GCM climate quadrants for the Berg WMA ......................................................................... 82
Figure 29: Distribution of average monthly ETm /PEm ratios ................................................................. 83
Figure 30: Comparison of historic monthly averages for potential reference crop evapotranspiration
over the study area calculated using the Penman-Monteith method over a 50-year period (1950 –
1999) and the Thornthwaite method over a 35-year period (1979 – 2013). .................................. 84
Figure 31: % increase in irrigated water requirements for Berg WMA in 2025 and 2040..................... 85
Figure 32: % increase in irrigated water requirements by municipality in 2025 and 2040 using
CANEMS2 45 climate model ......................................................................................................... 86
Figure 33: 2025 estimated irrigated water requirements by municipality ............................................. 87
Figure 34: 2040 estimated irrigated water requirements by municipality ............................................. 87
Figure 35: % increase in irrigated water requirements by crop in 2025 and 2040 for the Berg WMA.. 88
Figure 36: % increase in irrigated water requirements by crop in 2025 and 2040 for the Berg WMA
(excluding grains)........................................................................................................................... 89
Figure 37: WCWSS allocations by user type (GreenCape, 2017) ........................................................ 90
Figure 38: Urban water usage by municipalities in the Berg WMA....................................................... 92
Figure 39: Total urban water requirements per municipality in 2011 .................................................... 93
Figure 40: 2011 Per capita urban consumption per town (m3/capita/a) ................................................ 94
Figure 41: % change in urban water requirements in 2025 and 2040 per municipality according to
historical population growth rates .................................................................................................. 96
Figure 42: Water treated per year and population in the City of Cape Town ....................................... 96
Figure 43: Urban water requirement growth rates for the Berg WMA under different population growth
scenarios ........................................................................................................................................ 97
Figure 44: Urban water requirements per municipality in 2025 ............................................................ 98
Figure 45: Urban water requirements per municipality in 2040 ............................................................ 98
Figure 46: Total water requirements for the Berg WMA in m3/a ........................................................... 99
viii
Figure 47: Total water requirements by municipality in m3/a .............................................................. 100
Figure 48: Total annual water requirements by municipality .............................................................. 101
Figure 49: Comparison between urban and agricultural water requirements growth rates in 2025 and
2040 by municipality .................................................................................................................... 102
Figure 50: 2025 total annual water requirements by municipality....................................................... 102
Figure 51: 2040 total annual water requirements by municipality....................................................... 103
Figure 52: 2025 Value of water supply deficit in 2015 R millions ....................................................... 106
Figure 53: 2025 Value of water supply deficit in employment ........................................................... 107
Figure 54: 2040 Value of water supply deficit in 2015 R millions ...................................................... 107
Figure 55: 2040 Value of water supply deficit in employment ........................................................... 108
Figure 56: 2025 comparison of GVA deficit to 2015 total economy.................................................... 110
Figure 57: 2040 comparison of GVA deficit to 2015 total economy.................................................... 110
Figure 58: Schematic showing the different processes making up the tool ........................................ 113
Figure 59: Selection of alternatives ..................................................................................................... 113
Figure 60: Criteria definition ................................................................................................................ 114
Figure 61: Criteria Metrics ................................................................................................................... 115
Figure 62: Criteria positive or negative relationship ............................................................................ 116
Figure 63: Criteria worst case acceptable value ................................................................................. 117
Figure 64: Criteria best case acceptable value ................................................................................... 118
Figure 65: Criterion legend.................................................................................................................. 119
Figure 66: Criterion scoring scale ....................................................................................................... 120
Figure 67: Value function shapes ....................................................................................................... 121
Figure 68: Alternative scoring ............................................................................................................. 122
Figure 69: Weighting questions .......................................................................................................... 124
Figure 70: Implied weights .................................................................................................................. 125
Figure 71: Manual weights .................................................................................................................. 126
Figure 72: Alternative results .............................................................................................................. 127
Figure 73: WCIP alternatives .............................................................................................................. 127
Figure 74: Selected criteria for SBLM evaluation................................................................................ 128
Figure 75: SBLM criteria metrics ......................................................................................................... 129
Figure 76: Definition of criteria as positive or negative relationship.................................................... 130
Figure 77: Criteria best and worst case values ................................................................................... 131
Figure 78: Potable water use value function graph ............................................................................ 132
Figure 79: Potable water use scores .................................................................................................. 132
Figure 80: Potable water use project alternative values and scores .................................................. 132
Figure 81: Potable water use shape ................................................................................................... 132
Figure 82: Non-potable water use value function graph ..................................................................... 133
Figure 83: Non-potable water use scores ........................................................................................... 133
Figure 84: Non-potable water use project alternative values and scores ........................................... 133
Figure 85: Non-potable water use shape ............................................................................................ 133
Figure 86: Operational jobs value function graph ............................................................................... 134
Figure 87: Operational jobs scores ..................................................................................................... 134
Figure 88: Operational jobs alternatives values and scores ............................................................... 134
Figure 89: Operational jobs shape ...................................................................................................... 134
Figure 90: Potential for local jobs value function graph ...................................................................... 135
Figure 91: Potential for local jobs scores ............................................................................................ 135
Figure 92: Potential for local jobs alternatives values and scores ...................................................... 136
Figure 93: Potential for local jobs shape ............................................................................................. 136
Figure 94: Environmental impact value function graph ....................................................................... 137
Figure 95: Environmental impact scores ............................................................................................. 137
Figure 96: Environmental impact values and scores .......................................................................... 137
ix
Figure 97: Environmental impact shape ............................................................................................. 137
Figure 98: Weighting question responses ........................................................................................... 138
Figure 99: Criteria weights for base case ........................................................................................... 139
Figure 100: SBLM project alternative scoring results ......................................................................... 140
List of tables Table 1: Capacity and yield of the dams and system of the WCWSS (DWS, 2015a) .......................... 10
Table 2: 2014/15 Consumption versus allocation million m3/a (DWS, 2015a) ..................................... 11
Table 3: Towns served by the Withoogte bulk scheme (DWS, 2015a) ................................................ 13
Table 4: Generic IDP process (adapted from DRDLR, 2012). The dark green block indicates the point
at which sector development plans are incorporated into the IDP ................................................ 31
Table 5: Characteristics of key water and planning forums in the Berg ............................................... 43
Table 6: Strategic Goals of the SWMP ................................................................................................. 45
Table 7: Employment in agriculture for municipalities in study area in 2011 ........................................ 56
Table 8: Employment estimates for irrigated crops in the Berg WMA .................................................. 57
Table 9: Comparison of fruit sector employment estimates .................................................................. 58
Table 10: Labour multiplier comparison (jobs/hectare) ......................................................................... 58
Table 11: Average value of agriculture and urban water in municipalities ........................................... 63
Table 12: Proposed indicators for SDG 6.1, 6.2 and 6.3 in Berg WMA ................................................ 64
Table 13: Total employment and “job per drop” for agriculture and non-agriculture sectors in the
municipalities of the Berg WMA ..................................................................................................... 65
Table 14: Seasonal time-averaged crop coefficients (Kc) for each irrigated crop type ........................ 70
Table 15: SAPWAT4 Wine Grape growing season and corresponding crop coefficients .................... 71
Table 16: SAPWAT4 Wine grape crop coefficients by season ............................................................. 71
Table 17: Effective rainfall estimates based on accumulated rainfall using the U.S. Department of
Reclamation Method (Adapted from Adnan & Khan, 2009) .......................................................... 72
Table 18: Irrigation types found in the study area with relative efficiencies (DoA, 2013 & Ascough &
Kiker, 2002) .................................................................................................................................... 72
Table 19: Irrigation requirements database sample ............................................................................. 73
Table 20: Comparison of project estimations of irrigated water requirements at regional level to other
available models ............................................................................................................................ 79
Table 21: 2011 Town water usage and population figures ................................................................... 91
Table 22: Comparison between 2011 and 2016 population growth rates by municipality (Stats SA,
2016) .............................................................................................................................................. 95
Table 23: Total annual water requirements for the Berg WMA ............................................................. 99
Table 24: Existing water allocations to urban or industrial users ........................................................ 104
Table 25: Urban and agricultural supply deficit in 2025 and 2040 m3/a ............................................. 105
Table 26: Water supply deficit opportunity costs in 2025 and 2040 ................................................... 108
Table 27: Comparison of the opportunity cost from a water supply deficit to the current size of the
local economy in 2025 and 2040 ................................................................................................. 109
x
Acknowledgements
The authors would like to thank the following Reference Group and Steering Committee members for
their contribution to the study, through constructive discussions at group meetings:
Eiman Karar (Water Research Commission)
Lauren Basson (GreenCape)
Dawie Mullins (Conningarth Economists)
Haroon Karodia (Retired, ex DWAF, DEA)
Ashia Peterson (Department of Water and Sanitation)
Bonani Madikizela (Water Research Commission)
Joe Mulders (Prime Africa)
Fernel Abrahams (Provincial Department of Economic Development & Tourism)
Anzel Venter (Provincial Department of Economic Development & Tourism)
Herman Jonker (Provincial Department of Economic Development & Tourism)
Paul Hardcastle (Provincial Department of Environmental Affairs and Development Planning)
Jim Petrie (Independent)
Valentina Russo (University of Cape Town)
Theodor Stewart (University of Cape Town)
Chrizelle Kriel (Provincial Department of Environmental Affairs and Development Planning)
Bruce Raw (GreenCape)
Helen Davies (Provincial Department of Economic Development & Tourism)
Amanda Dinan (Fetola)
Masters students at the University of Cape Town, Marthinus Van Der Walt and Jacob Muller are
acknowledged for their contributions to:
Assessment of current and future water requirements in the Berg based on ‘bottom-up’
calculations of irrigation water requirements and collation of domestic water use datasets, and
Assessment of the marginal cost of water in the agricultural sector in the Berg, respectively.
Their master’s theses are referenced at the appropriate places within this document. Furthermore,
their supervisors Dr Jane Turpie, Anton Cartwright, Dr Julian Smit, Dr Nadine Methner and Professor
Mark New are acknowledged for their contribution in shaping the research underpinning this study,
and guiding the project team at team meetings.
In addition, Megan Cole’s contribution to the project as part of her doctoral thesis (Cole, 2017) is also
acknowledged as it provided data on the social value of water access in the study region.
Special mention for the contribution made by Dean van der Heever who generated all the maps in this
report, developed the GIS model for the regional hydro-economic tool and published interactive maps
for decision-makers to interact with on Mango Maps.
Water Research Commission (WRC) project K5/2453, “Towards Sustainable Economic
Development in Water Constrained Catchments: Tools to Empower Decision Making” was
awarded in response to an unsolicited proposal submitted in 2014. The project was co-funded by the
WRC and Department of Economic Development & Tourism (DED&T) of the Western Cape
Government (WCG).
xi
List of acronyms
ACDI African Climate and Development Initiative
BCC-CSM1-1 Beijing Climate Center Climate System Model
BFAP Bureau for Food and Agricultural Policy
BGCMA Breede-Gouritz Catchment Management Agency
BNU-ESM Beijing Normal University Earth System Model
BRIP Berg River Improvement Programme
CanESM2 Second generation Canadian Earth System Model
CBA Cost-Benefit Analysis
CCT City of Cape Town
CMA Catchment Management Agency
CMS Catchment Management Strategy
CNRM-CM5 Centre National de Recherches Météorologiques’ GCM
CBA Cost-Benefit Analysis
CSAG Climate System and Analysis Group
CUC Capital Unit Charge
DAFF Department of Agriculture Forestry and Fishing
DEA&DP Department of Environmental Affairs and Development Planning of the Provincial Government of the Western Cape
DED&T Department of Economic Development and Tourism of the Provincial Government of the Western Cape
DM District municipality
DoA Western Cape Provincial Department of Agriculture
DPLG Department of Provincial and Local Government
DRDLR Department of Rural Development and Land Reform
DWA Department of Water Affairs
DWAF Department of Water Affairs and Forestry
DWS Department of Water and Sanitation
EDP Western Cape Economic Development Partnership
EIA Environmental Impact Assessment
FIBC Future Infrastructure Build Charge
FGOALS-s2 Flexible Global Ocean-Atmosphere-Land System model, Spectral Version 2
GA General Authorisation
GCM General Circulation Model
GDP Gross Domestic Product
GFDL-ESM2G National Oceanic and Atmospheric Administration Geophysical Fluid Dynamics Laboratory’s Earth Systems General Model
xii
GFDL-ESM2M National Oceanic and Atmospheric Administration Geophysical Fluid Dynamics Laboratory’s Earth Systems Modular Ocean Model
GIS Geographic Information Systems
GS RSIF Greater Saldanha Regional Spatial Implementation Framework
GVA Gross Value Add
HDI Historically Disadvantaged Individual
IBT Inter-Basin Transfer
IDP Integrated Development Plan
IDZ Industrial Development Zone
KZN KwaZulu Natal
LED Local Economic Development
LM Local municipality
LUPA Land Use Planning Act
MCDA Multi-Criteria Decision Analysis
MERO Municipal Economic Review and Outlook
MIG Municipal Infrastructure Grant
MIROC5 Model for Interdisciplinary Research on Climate and associate Earth System Models
MRI-CGCM3 Japanese Meteorological Research Institute’s Coupled Global Climate Model
MTSF Medium Term Strategic Framework
NDP National Development Plan
NEMA National Environmental Management Act
NPC National Planning Commission
NWA National Water Act
NWRS National Water Resources Strategy
NWSMP National Water and Sanitation Master Plan
PICC Presidential Infrastructure Coordinating Commission
PSDF Provincial Spatial Development Framework
RBIG Regional Bulk Infrastructure Grant
ROA Return-On-Assets
RQO Resource Quality Objective
RSA Republic of South Africa
RWU Regional Water Utility
SATI South African Table Grape Industry
SAWIS South African Wine Industry Information and Systems
SBLM Saldanha Bay Local Municipality
SDBIP Service Delivery Budget Implementation Plan
SDF Spatial Development Framework
SIP Strategic Integrated Project
SPLUMA Spatial Planning and Land-use Management Act
xiii
SWMP Sustainable Water Management Plan
TCTA Trans-Caledon Tunnel Authority
UCT University of Cape Town
UWC University of the Western Cape
V&V Validation and Verification
WARMS Water Authorisation and Registration Management System
WC/WDM Water Conservation/Water Demand Management
WCDM West Coast District Municipality
WCG Western Cape Government
WCWSS Western Cape Water Supply System
WMA Water Management Area
WRC Water Research Commission
WRCS Water Resource Classification System
WRSM Water Resource Simulation Model
WSA Water Services Authority
WSA Water Services Authority
WSDP Water Services Development Plan
WSP Water Services Provider
WTE Water Trading Entity
WUL Water Use License
WULA Water Use License Application
WWAP World Water Assessment Programme
1
1. Introduction
1.1. Study Vision
Water resources planning and economic development planning need to acknowledge that the two
factors are inter-dependent. Currently, each one is treated as an independent variable in the
other.
1.2. Study Motivation
There is wide recognition that the combined effects of climate change, population growth and
continued urbanisation are exerting pressure on limited water resources. By 2012, approximately
half of the major South African supply schemes were already in a water balance deficit, requiring
new water resources interventions to meet projected future demands. According to the
Department of Water Affairs (DWA) (2013a), 55% of smaller schemes supplying settlement areas
or towns are currently or will be in a water balance deficit before 2023. At the same time economic
growth is vital to alleviate poverty, hence the large national drive to stimulate growth (NPC, 2011).
Given that growth is required in the face of natural resource constraints, the Green Economy has
been promoted most broadly as an approach to maintain growth whilst not depleting natural
resources (WCG, 2013).
In terms of water resources and development, the Department of Water and Sanitation (DWS)
has been careful to point out that whilst water is essential to development, its availability is not a
driver to, nor constraint on, development (DWA, 2009; DWA, 2010). This position of DWS is based
on the view that as much water can be made available as is required (via desalination for
example). In the case of a catchment where all readily available water is allocated (referred to as
a ‘constrained catchment’), future development requires additional water resources. This in turn
requires new infrastructure, which comes at a cost. This cost would be borne in part by the future
water users (people, economic developments), via capital levies or direct water charges. If the
users considered part of the future economic development are unable to bear this cost, then the
cost of provision of water becomes a constraint to economic development. Furthermore, the
capacity to develop this infrastructure may not always be available locally and is therefore a
stumbling block to many local municipalities. The ability of the users to carry the cost of this
proposed water infrastructure needs to be taken into account in determining the viability of the
considered future economic development (GreenCape, 2015). Furthermore, the security of
available water resources will promote investment in a region, in the same way that uncertainty
over future supplies, or the cost of ensuring water security, has in several cases dissuaded
investment. There is an interdependency therefore between economic development and water
resources, which needs to be taken into account in development planning (key intervention 1).
Given the potential constraint of (the cost of) water, allocation or access to water services in a
constrained catchment should be towards those developments that maximise environmental and
socio-economic benefits of the water used. Of course, economic benefit is challenging to quantify,
and there are complex links to considerations of socio-environmental benefits for water use or
allocation. But at the least, environ-socio-economic benefit considerations need to be
incorporated in water allocation or access decisions to promote the ‘smartest’ use of water (key
intervention 2). This is also promoted by DWA (2009).
2
It follows that in constrained catchments, there may simply not be ‘enough for all, forever’ and
regional planning decisions between potentially competing development trajectories need to be
taken (key intervention 3). For example, what is the impact on the economy of diverting more
water towards agriculture in a bid to promote food security? Conversely, what is the impact on
the food processing industry and on food security, of a decision to promote more economically
lucrative uses of water than agriculture? What economic developments should be promoted
regionally, and in what locations? In a perfect water market, market forces would dictate the
access to water between competing users. However, water is identified as a basic human right in
the South African Constitution giving priority to domestic use. Meeting the Ecological Reserve is
also a priority as dictated by the National Water Act (Act 36 of 1998). Also, water’s role in food
security ensures that high priority is given to agricultural use. Therefore water allocation and
access decisions have to be made while achieving sustainable environ-socio-economic growth.
The need for sustainable growth requires that various uses of water be assessed for their local
and regional costs and benefits.
1.3. Previous/Related Work
Driven by a similar motivation to that behind the key interventions highlighted above, a recent
WRC project investigated the links between water resources and the economy in the Western
Cape, and attempted to understand how water flows through the economy (Pegram and Baleta,
2013). The applicability of various tools for linking water and economics, such as virtual water and
indices such as rand per drop, were assessed. The project unpacked and promoted some critical
paradigm shifts required in order to assess water and economics as one linked package. Other
researchers are also applying complex systems thinking approaches to try to unpack water
resources planning and the inherently linked considerations such as economic development, and
management of the water-energy-food nexus (such as Muller, 2013 and Palmer, 2014). Another
research team completed a regional resource flow model to develop a baseline for the resource
efficiency of the Western Cape economy, benchmark sectors and particular commodities in these,
with the aim of identifying interventions to increase the resource efficiency of the various sectors
and the economy as a whole (Janse van Vuuren & Pineo, 2014). These studies improve our
current understanding of how water moves in the economy.
In addition, a model has been developed for demonstrating the importance of water in the South
African economy, while providing a means for quantifying the impact of different water policy
strategies and demand/supply scenarios (Conningarth Economists, 2012). The model allows
comparison of the benefits of water use between various sectors using weighted average
multipliers (GDP, number of employees, households) per Rand per m3 water used, and also
allows for growth scenario analysis and the aggregated effects of different policy interventions
such as increasing water tariffs. However, this existing model does not take into consideration the
full complexity of the system, to quantify trade-off or knock-on effects of different water uses. For
example, whilst demonstrating that agriculture uses water less ‘efficiently’ than other sectors, the
analysis does not take into consideration issues such as the value of food security and regional
imbalances in the prevalence of poverty. The model also excludes the environmental costs or
impacts of various proposed water uses or developments, and of various water resources
interventions. In addition, the model does not consider the spatial relationships between economic
growth and the variability of water availability and quality across the catchment. Finally, although
3
the model considers economic indicators (GDP and jobs contribution at a macro-scale) it does
not include social impact indicators such as changes in well-being.
As yet there are no known examples where planning that acknowledges the inter-dependency of
water and economics (key intervention 1 above) has been implemented. There are also no known
examples of cost-benefit analyses or socio economic-benefit of proposed developments,
informing current decisions over proposed developments, prior to investing in them (key
intervention 2 and 3 above).
1.4. Focus on Berg Water Management Area and Saldanha Bay
Whilst the interventions motivated above (key intervention 1, 2, 3) are theoretically relevant to any
constrained catchment, there is an urgent need to implement these interventions to inform
economic development decisions in the Saldanha Bay area, within the constrained catchment of
the Berg River. The West Coast District Municipality (WCDM) is supplied by water from the
Western Cape Water Supply System (WCWSS), via the Berg River and Misverstand Dam, and
in turn supplies water to Saldanha Bay Local Municipality (SBLM). The WCDM has exceeded its
allocation since 2008 (DWS, 2015). At the regional catchment level, allocations from the WCWSS
should weigh up the potentially competing demands of industrial development in Saldanha versus
agriculture in the Middle and Upper Berg, versus domestic and industrial supply to the Cape Town
metropolitan (Pegram and Baleta, 2013). Additional water resources need to be brought online
for Saldanha’s planned development, and have been investigated (WCDM, 2010), however are
yet to be implemented largely due to financial challenges. SBLM has previously declined
applications for large-scale access to water (forcing a proposed development to source their own
water supplies) and made access dependent on SBLM developing new water resources.
Economic growth in the region therefore might become constrained by the cost of and availability
of water (GreenCape, 2015).
In baseline research (GreenCape, 2015) and a series of interviews with decision makers in
economic development planning and water resources planning, the following challenges have
been observed for Saldanha Bay:
Water resources and economic development plans are generally each treated as
independent variables in the planning of the other. One takes the other “into account” but
does not consider the inter-dependency. This is entrenched by the planning protocols such
as Integrated Development Plans, Water Services Development Plans, and Master Plans.
This disconnect in the planning system has led to the current situation in which those
responsible for water resources allocation assume that industries with high water demand
should be ruled out for the area. However, consideration should be given to the possibility
that economic productivity from these industries and future water users may outweigh the
costs of the required water infrastructure. This can only be assessed if a systems approach
to planning is implemented, and if the total socio-economic-environmental cost/benefit of
development options or water resources allocations are assessed and used in decision-
making.
Some of those in the planning system recognise the current challenges and reflect that there
is no current alternative. Projects are currently assessed on an individual basis (i.e. in an
environmental impact assessment or a water use license application), rather than strategic
assessments of development scenarios. Local-scale planning depends on Provincial
4
Government for this strategic oversight role. At this level, the full spatial complexity of the
linked socio-economic-resources system needs to be taken into account. However, there
are currently no tools to assist in this assessment.
Palmer et al (2014) noted similar challenges when unpacking the reasons for slow implementation
of the National Water Act, and motivates that a new paradigm is required for water resources
planning and associated development planning, which incorporates a linked systems approach
to assess trade-offs between competing uses (Palmer et al, 2014).
1.5. Research aims and proposed tools
Based on the evolving vision and the challenges identified, the project proposed three potential
solutions:
1. The development of a cost-benefit analysis of proposed economic developments, water
allocations, and resource interventions (key intervention 2)
2. The development of a regional hydro-economic model (key intervention 3 above),
3. An investigation into the current state of water resources and economic development
planning to guide the development of the tools (item 1 and 2 above) inform how best to
implement the tools, to promote better integrated planning (address key intervention 1
above)
The ‘solutions’ are indirect measures that aim to alleviate the potential constraint from poor
planning, and enable smarter decisions necessary with constrained resources. The project did
not assess the various direct solutions available to alleviate water resource constraints in
Saldanha and the Berg WMA, i.e. provision of new water resources.
The overarching intention of the project was to (through the tools to be developed, and through
the process of their development) inform decision-making for economic development and water
resources, promote or take steps towards change in the current approach to planning, and ensure
research findings or proposed tools were implemented during the course of the project. There is
a growing recognition that to achieve this intention, research must be co-created by those it is
intended to benefit (i.e. in this case decision-makers), at all stages including even the inception
of such research, to ensure it is fit for purpose and is implemented. As such, the project team
fostered close working relationships with decision-makers, directly connecting them with
academic research.
Although the proposed research responds to the challenges in Saldanha Bay and Berg WMA,
these challenges and the proposed interventions listed above are common to other constrained
catchments, hence the methods and lessons from this case study are transferable to other areas.
1.6. Study spatial boundaries and focus areas
The study area includes the (previous) Berg Water Management Area (WMA), shown in Figure
1. This WMA has since merged with the Olifants WMA to the north, and forms the Berg-Olifants
WMA. However, in this document, the term Berg WMA is used and refers to the previous outline
of the WMA. The Berg WMA comprises the Berg River catchment together with areas along the
west coast including the Diep River catchment and the Greater Cape Town area.
The Berg WMA includes a number of local municipalities, as well as the City of Cape Town
metro. See the map below that delineates the location of the municipalities within the WMA.
5
Figure 1: Berg Water Management Area including municipal boundaries
This region was the focus of the study, and wherever possible the boundary of the WMA was
the extent of the study analysis. The exception to this was during some of the economic
analysis of the local municipalities, where those municipalities fell across the boundary but
where it was difficult to distinguish the specific economic outcomes for the proportion that fell
within the WMA. Additionally, SBLM, was the focus for the “local tool” intervention, and many
engagements1 were held within the municipality to assist with the water constraints being
experienced and the impact that this was having on the development potential of the area.
The analytical focus area of the study was largely concerned with the impact of water availability
(i.e. not water quality) on the economic outcomes of the region (i.e. jobs and investment). The
assessment of the planning systems looked into the intersection of development planning, water
allocation and water resource infrastructural development. The economic analysis valued water
on the basis of Gross Value Add (GVA) and employment. The calculation of water requirements
was based on water access for production and human consumption. Therefore the impact of
water availability on the environment was not included in the analysis.
1 For example, the project team assisted with the scoping of the “Water Exchange Network”, regularly met
with industry players to understand their water requirements and growth plans, and assisted the
municipality as part of the municipality’s “Smart Water Team”.
6
1.7. Outline of this document
The document follows the following structure:
Chapter 2 discusses the methodological approach to the study
Chapter 3 provides background and context to the study area
Chapter 4 analyses the planning and governance approaches to water allocation, water
resources and economic development
Chapter 5 discusses the role that water plays in the regional economy and provides some
estimates of its value
Chapter 6 estimates how much water is being used in the region, and forecasts how much
will be required in 2025 and 2040, as well as highlighting the potential gaps in supply with
the implications for development
Chapter 7 discusses the tool that was developed for the SBLM to assist in prioritising water
applications
Chapter 8 summarises the key insights and findings from the study
Chapter 9 provides recommendations for future research and the way forward
Thereafter references are provided, as well as a number of appendices with details from the
research
7
2. Methodology
2.1. Project approach
The solutions developed on the project include two tools (a regional hydro-economic GIS model,
and a local development prioritisation tool), and a governance assessment (or planning
guidelines) to guide development and direct implementation of the two tools. Figure 2 shows the
diagrammatic representation of the overall project structure:
Figure 2: Project structure
The diagram illustrates how there were two distinct, yet related project areas: Saldanha Bay and
the Berg River Catchment. There were a number of multi-disciplinary applied research studies
that were consolidated into the Regional Hydro-Economic GIS tool – these were broadly split into
estimating the total water requirements and then valuing those requirements. The Hydro-
Economic tools were underpinned by the planning guidelines or governance research.
GreenCape was positioned as the coordination hub.
2.2. Co-production with decision-makers
The overarching intention of the project was to (through the tools to be developed, and through
the process of their development) inform decision-making for economic development and water
resources, promote or take steps towards change in the current approach to planning, and ensure
research findings or proposed tools were implemented during the course of the project. To
achieve this aim, and ensure that the solutions meet the requirements of the decision-makers that
will need to implement them, close collaboration and co-production with decision-makers was
sought throughout the project. Formal workshops, smaller group meetings, and one-to-one
meetings were held regularly. GreenCape also participates on several steering committees of
similar projects which provided a platform for discussion and collaboration.
8
Following on from the scoping study, the proposal to the WRC for this project contained a problem
definition and proposed solutions (outlined in section 1). Preliminary investigations into necessary
datasets, and research into the proposed solutions, was also carried out during proposal phase.
At the initiation of the project, a “user-needs workshop” was held to verify the problem definition,
and shape solutions. This engagement led to the prioritisation of the challenges detailed in the
problem definition, and shaping of proposed solutions, further data collation and tool development
(further detailed in section 3.3). At routine interactions, the research was shared with decision-
makers, and course-corrected where necessary, and the interactive cycle was therefore repeated
several times (summarised in Figure 3). Continued interaction with decision-makers ensured that
the project team remained informed of the problem landscape, as this changed significantly over
the years since the proposal was written. In some cases, elements of the initial proposal were no
longer valid, yet it was broad enough to allow the tools to be shaped accordingly. Particularly, the
severity of the 2015-17 drought in the Western Cape certainly heightened interest in water, and
provided GreenCape with additional platforms to disseminate research findings and foster
implementation of the developed tools, although most decision-making during the period was
being driven by the crisis and not longer-term planning.
GreenCape has supported decision makers in implementing the tools from this project through
technical guidance and facilitation services. Implementing solutions and interventions was a
continual process and was not reserved only for the end of the project.
In addition to closely involving decision-makers in the project, the research process itself
represented a cross-sectoral collaboration between academic research housed at a university,
consultants, a non-profit organisation, and (local, provincial, and national) government decision-
makers.
Lessons learnt in the project process, and an appraisal of the successes of the project are
included in section 8.3. In addition, a record of the liaison with decision-makers and research
champions with whom the project has engaged with during the course is included in Appendix 1.
Three annual workshops were held on the project, and their findings are integrated into the report
in the relevant sections, with further details given in Appendices 2 and 3.
9
Figure 3: Framework for the iterative, co-development of solutions with decision-makers
Problem defintion
Shaping of proposed solutions
Collection of data and relevant information
Desk-top research for developing solutions
Liaison, verification &
testing by decision-makers
Implementation
of ‘final’ tools
10
3. Study Area Background and Challenges
3.1. Water resources in the Berg Water Management Area
The Department of Water and Sanitation (DWS) previously divided the country into 19 Water
Management Areas (WMAs), each containing a large river system (DWAF, 2004a). The Berg
River catchment supplies areas outside of its natural boundaries (Cape Town for example), and
the boundary of the Berg WMA includes the supply area, and several smaller catchments. With
the second revision of the National Water Resources Strategy (DWA, 2013) 19 WMAs were
reduced to nine, through an amalgamation of areas. As such the Berg no longer constitutes an
individual WMA, and is now part of the Berg-Olifants WMA, but is still referred to as such in this
report.
The Berg WMA is a heavily utilised system. Through a set of six major dams (listed in Table 1),
the Berg River supplies water to the majority of municipal demands (both domestic and
industrial) as well as significant agricultural demands across the Berg WMA. The structure of the
supply network, collectively termed the Western Cape Water Supply System (WCWSS, Figure
4), enables these dams to be operated in an integrated manner, maximizing efficiency. One of
the dams (Theewaterskloof) is located in the neighbouring Breede River catchment, and
contributes water to the Berg Catchment and the WCWSS via an inter-basin transfer. The DWS
(2015a) points out that the total integrated system yield (at a 98% level of supply assurance for
all user categories) is 570 million m3/a. This figure is deduced from the maximum annual water
requirements that the system can supply before the risk of curtailments become too great.
Table 1: Capacity and yield of the dams and system of the WCWSS (DWS, 2015a)
Dam Capacity [million m3] Yield [million m3/a]
Theewaterskloof Dam 432 219
Voelvlei Dam 158 105
Wemmershoek Dam 58 54
Upper Steenbras Dam 30 40
Lower Steenbras Dam 34
Berg River Dam 127 80
Palmiet 23
Compensation 38
Additional yield from integration 11
Total 839 570
11
Figure 4: Western Cape Water Supply System network infrastructure
The WCWSS supplies raw water to domestic and industrial users (via municipal supply), and to
agricultural users (via Water User Associations and irrigation boards). The major domestic and
industrial user is the City of Cape Town, which then supplies to users within the Metropolitan
area. The West Coast District Municipality (WCDM) also receives water from the WCWSS and
operates smaller schemes to supply West Coast towns and industry, as does the Stellenbosch
Local Municipality. The latest WCWSS consumption and allocation figures are from 2014/2015
and these reveal that this system was at that stage already over-allocated (i.e. that the water
use authorisations exceeded the estimated system yield), however the usage was not yet
exceeding the yield of the system:
Table 2: 2014/15 Consumption versus allocation million m3/a (DWS, 2015a)
Allocations 2014/15 Use
City of Cape Town 357.90 334.70
WCDM 22.80 27.73
Stellenbosch 3.00 4.19
Other urban/industrial 9.18 10.64
Agriculture (capped) 216.24 170.00 (estimated)
Total 609.12 547.26
12
The DWS planning scenario (based on high water requirement growth, 50% success of water
conservation and water demand management measures and no impact of climate change)
indicated that the WCWSS’s water requirements would exceed the system yield in 20192. The
first possible supply augmentation scheme (Voëlvlei Augmentation Scheme) will increase the
system yield by 23 million m3/a and may come online in 2021 (due to the current drought, this
scheme has been prioritised and may be online earlier than previously reported). However, as
this brings the system yield to 605 million m3/a, the system will still be over-allocated. Thereafter
a number of new supply schemes will need to be implemented in to meet the continued growth
demands of the system. Feasibility studies are underway by the City of Cape Town (CoCT) for
large-scale desalination, water reuse and groundwater use, and implementation of one of these
schemes would have to commence imminently. Short-term schemes are being planned by the
City of Cape Town that are effectively piloting the Table Mountain Aquifer Group and large-scale
reuse schemes, yet also form part of the City’s drought emergency supply schemes.
Figure 5: WCWSS reconciliation of supply and demand for the Planning Scenario (Redrawn
from DWS, 2015a)
The WCWSS is therefore highly constrained according to existing allocations (which are
undergoing a Validation and Verification process (see Box 1 in Chapter 4) and may therefore be
updated), with a high reliance on surface water resource. New supply options largely rely on non-
surface augmentation (excluding Voëlvlei) and are therefore expected to be far more expensive
to develop than previously built dams.
2 Sooner than previously thought due to decreased system yield
13
3.2. Water resources in Saldanha Bay Local Municipality
The total allocation to the WCDM from the WCWSS is 22.80 million m3/a, of which 17.4 million
m3/a is allocated from the Withoogte Scheme which supplies SBLM. WCDM can also extract 1.5
million m3/a from the Langebaan Road aquifer. The WCDM has exceeded its WCWSS allocation
for the last 8 years, and Withoogte for the last five years (see Figure 6 below).
Figure 6: Abstraction versus allocation to the Withoogte Scheme from the WCWSS (DWS,
2015a)
The WCDM applied for a water use licence in 2011 that would increase the allocation from
WCWSS to Withoogte, to 30.3 million m3/a by 2033. However, this increased allocation is in
competition with a similar application from the CoCT. This issue has not yet been resolved and
the application request has not been approved nor denied by DWS3. However, a recommendation
has been provided to DWS that the WCDM application is not awarded until other resource
interventions are in place for Cape Town, i.e. in 2021 (DWS 2015a).
Table 3: Towns served by the Withoogte bulk scheme (DWS, 2015a)
Bulk System Local Municipality Towns
Withoogte
Saldanha Bay Hopefield, Langebaan, Vredenberg, Saldanha, St Helena Bay
Swartland Koringberg, Morreesburg
Bergrivier Velddrif, Dwarskersbos
Of the approximately 21 million m3/a of water utilised in the Withoogte scheme, around 13.5 million
m3/a is sold from the WCDM to the SBLM for distribution, the remainder being sold to other LMs
from the Withoogte (or comprising losses).
3 As this study was going to print, confirmation was received by the WCDM that their licence application
had been approved, but no further details were available to the project team
14
Due to the delay in the increased allocation from the WCWSS, even if the WCDM application
were to be approved in 2021, on the basis of existing abstraction rates and potential demand from
new developments in the area, the increased supply from the WCWSS would not be sufficient to
meet projected demand in SBLM (see Figure 7 below).
Figure 7: Projected SBLM water allocation and abstraction (DED&T, 2016; DWS, 2015a)
The situation highlights the precarious position of the SBLM: it is a relatively small user within a
large scheme (in terms of volume used, population dependent on, and also GDP generated with
this water), and dependent on decisions made upstream of it.
Besides increased allocation from the WCWSS, the WCDM, on behalf of the SBLM, has also
pursued desalination as a water resource option. This option is favoured by the municipality over
other local resource development options as it has the highest assurance of supply. The proposed
project aimed to build a 25.5ML/d by 2026 at a proposed cost of R450 million (Weslander, 2013)
(current cost is estimated to be closer to R600 million). However the application to DWS for
Regional Bulk Infrastructure Grant funding was not sufficient to cover the total cost of the project,
and the municipality struggled to find funding for the shortfall. Additionally, DWS were concerned
that desalination had been prematurely favoured without adequate appreciation of alternative
sources, including increasing the supply from groundwater, and water reuse (GreenCape, 2015).
The plan for this desalination plant has therefore been shelved for the time being.
It seems likely that SBLMs lack of water availability will continue for the coming short-term (four
years) at the very least. Without a response forthcoming from DWS on the WCDMs water use
license application for increased allocation from the WCWSS, there is a lack of certainty as the
best way forward for assuring future supply. In the current drought environment, the need for
increasing drought resilience through diversification of supply has also been highlighted and
temporary local resources are being explored, including groundwater and reuse.
15
3.3. User Needs summary
The following six statements form a summary and grouping of the challenges listed in chapter 1,
that were identified through engagement with decision makers during the scoping phase
(GreenCape, 2015). They represent a mixture of challenges in water supply, and also challenges
in coordinated development planning. There is overlap between each of them.
1. Water demand may outstrip supply in 2017-2018
2. A misalignment in planning approaches, makes it difficult to strategically assess a set of
development options, and to know whether there is sufficient water to support these
3. There is no feedback loop between water demand, intervention cost, and whether the
development can support the intervention cost, and ultimately the development approval.
4. Although work has been done on the economic productivity of various water uses or
economic sectors, this had not yet been used to inform the allocation of water resources,
nor in what development scenarios should be promoted for Saldanha
5. Projects are awarded on project-by-project basis, without strategic oversight and
quantification of competing resource demands/trade-offs. Tools to enable this are lacking
6. The building block is missing: a coordinated picture/repository of planned development
These identified challenges led to the key interventions (or solutions), contained within the
proposal for this project, and summarised in chapter 1.2. At project inception, a half-day “user
needs workshop” was held in June 2015. The workshop was externally facilitated and 29
participants attended.
The purpose of the workshop was to align proposed project solutions to the needs of those
involved in water resources and economic development planning, and those impacted by this
planning. The first aim of the workshop was to verify whether these challenges are relevant,
critical, affecting development, and whether there were any missing. Once a collective picture of
challenges could be established, the second aim of the workshop was to collect stakeholder
feedback on the proposed solutions; to identify whether these solutions would add value, whether
there is buy-in for them, shape them to be most relevant, and identify decision-makers who would
be most likely to adopt the research outputs, and play key roles in the project process. The
participants were asked:
1. (How) Are these challenges affecting you? Can you prioritise them?
2. Are there any key (water & economic development) challenges missing?
Participants responded to the prioritisation question by placing each challenge in their perceived
relative order of importance (from 1 to 6 for 10 identified challenges, Figure 8). The results are
shown in Figure 9.The key results of this exercise show:
Although people considered it necessary to highlight four additional challenges, one of these
received no priority “votes”, and a further two received the fewest votes. This suggests the
dominating challenges were captured by the scoping phase, listed as 1 to 6 above.
Participants may have felt the need to raise issues that are linked, but these appear to not
be central to the group present.
16
The challenge that stakeholders feel is the most relevant, or biggest challenge for water
resources and economic development in Saldanha4, is that there is a misalignment in the
planning approaches.
The second biggest challenge for water resources and economic development in Saldanha,
is that there is a lack of tools for strategic oversight, followed by the lack of feedback loop
(between planned development, water demand, cost of water resources interventions,
whether the development can afford this cost and whether development should proceed).
The prioritisation exercise confirmed the worth of developing tools that can better integrate water
resources planning and economic development and assessing planning procedures in order that
these tools can be implemented.
The following two questions were posed to the participants to direct their discussions on the
proposed solutions:
Would these solutions be effective for you?
Can you see a way to improve the proposed solution?
There was support from participants (at the workshop as well as at other initial presentations) for
the proposed solutions.
A key theme from the user-needs workshop was the perceived limitation in understanding how a
full range of planning ordinances, policy directives and local by-laws, all operating at slightly
different scales and with different levels of authority, could be responsible for directing the water
resource and economic development of Saldanha and the Berg WMA. Workshop participants
highlighted the need for the project to begin by identifying the status quo of water resource and
economic development planning. This status quo was considered necessary to also shape tool
development further, and inform where how and by who the tools being developed could be
implemented.
4 Saldanha Bay and its development future drove the motivation behind this study, and the focus on the
Berg WMA emerged later as it was determined that Saldanha’s future was intricately linked to the
upstream water users in the remainder of the Berg WMA.
17
Figure 8: Stakeholders prioritising identified challenges (shown on white posters) through
voting (coloured numbered notes) in June 2015
Figure 9: Stakeholder perspectives on the relative importance of challenges identified
The following chapter outlines the assessment undertaken of the planning approaches that led to
the misalignment in Saldanha Bay and the Berg WMA.
18
4. Status quo of water resources and
economic development planning
4.1. Introduction
Integrated (water resources and economic development) planning requires that: i) legislation and
legislated planning processes support this integration; ii) that the necessary platforms are
available (and functioning) to support information exchange and integration; and iii) that
knowledge, data and information are available to support this integration. The user needs
summary (section 3.3) highlights the key challenge for integrated development is a misalignment
in planning, and a lack of tools. It therefore suggests that there are shortcomings in legislation,
and in the various platforms for planning, as well as a lack of knowledge, data and information.
The legislation that controls the way in which water is accessed, economic development is
planned, and water services and resources development is planned, was investigated. Current
forums responsible for water resources and economic development planning were also assessed.
The purpose of this assessment was:
To deepen the understanding of the status quo for water resources and economic
development planning, and thus understand the source of the challenges raised.
To guide the development of the various tools,
To identify where, how and by whom, the tools (and more broadly, the recommended
approach to integrated water resources and economic development planning) should be
used.
These investigations were carried out largely through review of relevant literature, policies and
relevant legislation for water resources and economic development, and one-on-one discussion
and workshops with people responsible for implementing the legislation or the forums. The
planning processes outlined in Figure 10, Figure 11, and Figure 13, were drafted and presented
at the second large workshop held on the project (June 2016). Stakeholders provided feedback
on how the processes operate in legislation and in practice (where this differs). This section
incorporates this feedback. Additional information on the second workshop is contained in
Appendix 2
The planning process findings have been separated into perspectives and practices (section 4.2),
current legislation (section 4.3) and current forums (section 4.4), and understandably there is
overlap between these sections.
4.2. Perspectives and practices
The below set of perceptions and practices have been collated through one-on-one interviews,
informal meetings and discussions, and annual project workshops with decision makers in
economic development planning and water resources planning. Findings from this assessment
expand on findings from the scoping phase appraisal of perspectives and practices water
resources and economic development planning, outlined in section 1 (particularly section 1.2) and
section 3 (particularly section 3.3).
19
The inter-dependence of water and economic development planning is not recognised in practice,
controlled in part by legislated approaches. Water resources and economic development plans
are generally each treated as independent variables in the planning of the other.
High water-demand development is often planned without assessment of available water
resources, or at least without assessment of the ability of the associated future water users
to fund water resources development. For example;
The establishment of the IDZ in Saldanha Bay assessed projected water
requirements availability, and made the assumption that desalination of
seawater was a potential resource and hence proceeded with the IDZ
declaration. The water requirements of the IDZ itself is not significant, but as
discussed in section 3.2, Saldanha Bay is already using more water than
allocated, without clear plans for future supply. If desalination is going to be
the approach pursued by the municipality to assure supply to the area, this
would be a substantial cost for future water users.
The WCG declared oil and gas and agri-processing as two of the priority
areas for economic development under project Khulisa (WCG, 2015).
However these are both water intense industries, effectively competing for
water within the Berg WMA. There was inadequate appreciation of the need
to develop regional water resources to support these industries, if they were
to achieve their objectives of job and investment growth (WCG, 2015)
On the other hand, water-intensive development is often disregarded under the assumption
that it will not be feasible in a water-constrained area, again without assessment of the ability
of the associated future users to fund water resources development. For example, DWS
made comments at a GreenCape hosted a workshop in February 2014 (scoping phase,
GreenCape 2015), that high water demand industries should be discouraged for Saldanha.
Water resources development plans are based on broad assumptions of the annual growth
in water requirements (as a percent) over the planning period (usually the coming 20-40
years). These water requirement growth percentages are generally developed through an
assessment of previous water requirement growth trajectories, and take into consideration
the general level of economic growth intended for a region, as documented in the Integrated
Development Plans (IDPs). Water resources interventions are recommended to meet any
future shortfalls, and costs provided. However, the IDPs do not outline specific private
investments and developments that are possible for the municipality (often unknown), and
are generated only on a 5-year basis. There are good reasons for these existing
approaches, however, it results in situations where potentially unaffordable infrastructure is
recommended (infrastructure that cannot be supported by the future users), or in a situation
whereby future potential development is not provided for.
Those responsible for water resources development planning reflect that there is a lack of
information on economic development planning, and as such are forced to make broad
assumptions for future water requirements.
The inter-dependence of water and economic development planning is recognised in perceptions
to some degree, or at least by some individuals. However, incorporating the inter-dependence
between water and economic development when implementing the required and legislated
planning processes is a challenge, and a lack of tools prevent integrated analyses.
Although work has been done on the potential future cost of water, and the cost per resource
intervention or source of water (desalination, water re-use etc.), and on the impact that this
cost may have on certain economies (Conningarth Economists, 2012; DWS, 2015b; DWS,
20
2015a; WCDM 2010), there is no work that addresses the complete cycle for Saldanha:
what are the potential developments, their water demand and the potential source, the
potential cost of the provision of the necessary water, and the impact of that cost on the
potential development.
Projects are currently assessed on an individual basis (i.e. in an environmental impact
assessment or a water use licence application), rather than strategic assessments of
development scenarios. This is also a legislative requirement; if a water use licence
application meets the required criteria it cannot be declined in favour of an application not
yet submitted which may reflect development that is more socially – economically
favourable. There does not appear to be an explicit feedback loop between the proposed
economic developments, the associated water requirements, the cost of water resources
intervention, and a strategic decision as to whether this proposed development, or group of
developments, is therefore viable.
Local-scale planning depends on provincial government for the strategic oversight role
required to meet this integrated regional planning (the vision, key intervention 1, section
1.2). At this level, the full spatial complexity of the linked socio-economic-resources system
needs to be taken into account. However, there are currently no tools to assist in this
assessment.
4.3. Legislated planning processes
4.3.1. Water allocation
4.3.1.1. Overview
Figure 10 represents a flowchart illustrating how water is allocated for use as dictated by the
National Water Act (NWA) (Act No. 36 of 1998, Section 21(a), taking water from a water resource).
The total water resource is determined through water resources yield models, and the allocable
resource is what is remaining after legal obligations are met (left-hand side blue boxes in Figure
10). These obligations include water for the ecological reserve, water for strategic use (for
example, electricity generation), international agreements and existing lawful use (light green
boxes in Figure 10). Remaining water that is allocable by the DWS or Catchment Management
Agency (CMA) includes all raw water resources (surface and groundwater) but excludes other
sources such as the use of seawater (via desalination).
Whether the DWS/ CMA acts as the responsible authority for allocation of the remaining water
depends on the catchment (responsible authority in dark grey in Figure 10). The NWA provides
for a water to be allocated by the relevant CMA, but since only two are currently operational, the
DWS remains the allocating authority in those catchments where CMAs are not yet established.
This remains an uncertain situation as there are indications that no more CMAs will be declared,
and that there will be a single, national CMA with regional/catchment offices that report to the
single CMA – however this will require changes to legislation.
Aside from the initial legal obligations, all other allocations from DWS function as a licensing or
authorisation processes, i.e. all taking of water from a water resource (i.e. direct users of raw
water) apply to DWS (or the CMA) for a licence to abstract water (dark grey boxes in Figure 10).
This includes water services authorities who must apply to DWS to abstract raw water, which they
21
then provide to their municipal domestic or industrial users (light grey boxes in Figure 10). The
process is controlled by Chapter 4 of the NWA. Water allocation priorities are dictated by Section
27 of the NWA, and are clearly mapped out in the second National Water Resources Strategy
(NWRS) (DWA, 2013a), and will also be described in the relevant Catchment Management
Strategy (CMS), where applicable. But it emerged through stakeholder engagement that these
priorities were not used in practice, and that the operational aspects of awarding WULs do not
easily allow for the realisation of these priorities.
The above description of Figure 10, and indeed the water use process described in the NWA, are
all straightforward. However, potential water users have great uncertainty over the allocation
process and the amount of water available for allocation. This uncertainty extends to within the
(water and economics) planning sector, the uncertainty arises from several sources such as lack
of information on how much water is available for allocation (and how this may change due to
climate change), and are related to slow implementation of all aspects of the NWA. CMAs are not
fully established, causing confusion, there is uncertainty over current water use, related to
incomplete registration of existing lawful use, and the ecological reserve is in many cases not fully
implemented, which would reduce the allocable resource once it is implemented. Some of these
complications, that have relevance to integrating water resources planning and economic
development planning, are described in further detail in section 4.3.1.2 and 4.3.1.3.
22
Figure 10: A flowchart of the water allocation process based on analysis of legislation
and verification by stakeholders
23
4.3.1.2. Total Allocable Resource
The calculation of the total water resource is a technical process that utilizes hydrological and
hydrogeological models, usually completed at primary catchment scale. The NWRS must
establish water management areas, contain a water balance for each area, provide the minimum
requirements for the reserve, set out actions to reconcile requirements and resources, set up
principles regarding WC/WDM and determine the inter-relationships between institutions. Once
the total water resource has been determined (through the NWRS), the NWA and the NWRS
highlight binding legal allocations (priorities) that must be reserved from the total water resource
(DWA, 2013a). These include:
The Reserve
According to the NWA (Section 16), the highest priority is afforded to water
for purposes of the reserve (RSA, 1998). The reserve has two components:
basic human needs and ecological reserve. The first objective is to ensure
that there is sufficient raw water available to provide for the basic water
needs of people, currently determined as 25 litres per person per day. The
second objective is to allow for the healthy ecological functioning of the
aquatic ecosystems through the implementation of the ecological reserve (a
technical process to determine the flow rate and quality required). These are
the only two absolute rights contained within NWA. The reserve is
determined (essentially a sign-off process) by the Minister. The placement of
the Reserve above the Total Allocable Resource in Figure 10 refers to the
non-negotiability of the implementation of the reserve before any allocations
can be considered. However, the human reserve component is not actually
reserved or set aside, and the ecological reserve is not currently
implemented in all areas.
International Agreements.
South Africa has committed to managing shared river basins in line with
existing international agreements with the relevant riparian states. These
include for example ensuring a particular flow transfers across borders. This
is not relevant in the case of the Berg area as it does not share its river basin
with any other countries.
Strategic Water Use.
Strategic water use refers to water use that is strategically important to the
national economy as per the NWA and includes water for inter-basin
transfers (IBTs) and electricity generation. IBTs are relevant for the Berg
insofar as the catchment receives IBTs from neighbouring catchments, but
has not committed to transfers out of the catchment. Freshwater
consumption for electricity generation is not significant within the Berg
catchment, as Koeberg is largely reliant on seawater, and the net water
consumption at Palmiet pumped storage scheme is low.
Existing Lawful Use
The NWA recognises historical water use as existing lawful use for end
users (Section 32). These lawful uses relate to (a) water used during an
interval period between the first democratic election in 1994 and the
commencement of the NWA in 1998 and (b) water used as a consequence
of riparian rights granted to farmers in the Water Act 54 of 1956. (RSA,
1956). In terms of (a) with the passing of the NWA in 1998, existing lawful
24
use of water was recognised and secured as a transitional measure until all
users could be licensed under the same regime (RSA, 1998). As such, DWS
are legally mandated to provide this water prior to determining the total
allocable resource. However, DWS are in the process of addressing this and
during a parliamentary portfolio meeting on the 2nd of November 2016, it was
emphasised that plans are in place to do away with this sunset clause5 in an
amendment of NWA. The validation and verification of the existing lawful use
is currently being conducted in the Berg and should be completed by the end
of 2017 (see Box 1). Following this process, DWS may implement
compulsory licensing to initiate a reallocation of water for the Berg River
catchment.
Once the above described legal obligations have been prioritised, the total allocable resource
available to a CMA or regional DWS office for allocation should be determined. According to the
NWA (RSA, 1998), the Minister is responsible for determining the quantity of water that the
responsible authority may allocate, subject to the guiding ‘priorities’ of the NWRS. These
‘priorities’ simply inform the manner in which the CMA/DWS reviews water use license
applications. In this sense, water allocation from the allocable resource functions as a licensing
or authorisation process rather than as a process driven by strategic priorities. This understanding
has been validated by a number of officials, including DWS regional office licensing officials and
the Breede-Gouritz CMA (BGCMA). Priorities are only considered when competing applications
are received and the key priority that governs that process is whether a minimum of 30% of the
allocated water will go to, or benefit, historically disadvantaged people. It is not clear however,
how completing applications from Water Service Authorities are compared, and there are
competing applications for water from the Berg WMA between Water Service Authorities (DWS,
2015a).
There are three types of (or levels) of water authorisation: Schedule 1, General Authorisation, and
Licences. Before applying for the relevant authorisation, the applicant will need to determine what
category they fall into.
Schedule 1 use refers to water for the purpose of household use only and is thereby
assumed to be low (although no specific usage rate is applied to it, rather only the purpose
of the water use). In this case, no application for a licence needs to be made, but Schedule
1 users are required to register with DWS to protect their interests.
General authorisations (GA) were designed to lighten the administrative load of licensing,
and are for volumes of water that fall within a range specified per quaternary catchment6 by
DWS. In known areas of water stress, the general authorization may be set extremely low
to enforce more users to require a full assessment that is required for licensing. The user
thereby does not apply for a (full) licence but must register their water usage under the
general authorisation category.
Water uses not falling into Schedule 1 or GA require a license, acquired through a water use
license application (WULA) process. The application is assessed by DWS regional office
(for the Berg), but signed off by the minister (i.e. national office). This transfers to the relevant
CMA once they are up and running. Presently, the BGCMA and the Inkomati CMA are the
5 A sunset clause refers to a piece of legislation that has an expiry date associated with it. In the case of
existing lawful use recognised during the ‘qualifying period’, this right will be lost following the completion
of the validation and verification process and compulsory licensing. 6 The limits for GA in the Western Cape are displayed here: http://mgo.ms/s/ll4ph
25
only two in operation in South Africa. The Berg-Olifants is a Proto-CMA so this responsibility
will transfer to the CMA once it has been fully established and the responsibility to issue
licences delegated from minister to CMA (if indeed this process is completed).
“Considerations” for the issuing of GAs and WULs are listed in Section 27 of the NWA, yet these
are not provided in section 27 in a prioritized order. The need to redress the results of past racial
and gender discrimination is listed as a consideration however no guidance is available in the
NWA or NWRS on how this is defined or implemented. The BGCMA has implemented this
criterion through requiring all applicants to have a minimum of 30% equity share for HDIs (pers
comm, Breede-Gouritz CMA, 12 October 2016). This has, however, led to legal disputes in other
areas.
A recent appeal court case of Makhanya v Goede Wellington Boerdery (Pty) Ltd,7 considered the
prioritisation of the considerations under section 27 of the NWA. The applicant, Goede Wellington
Boerdery, was refused a licence for the transfer of water rights from a neighbouring farm on the
basis that they had not met the equity consideration in terms of the criteria laid out in section 27
of the NWA. The court ruled that the transfer of water rights is possible, that an equal
consideration of various factors must take place in the consideration of any application for the
transfer of water rights, and that no single factor (out of those considerations listed in NWA
Section 27) may be afforded preference or carry a heavier weight than any other factor in such
consideration. The judgement confirms that, although empowerment is an important factor in the
consideration of licenses, it is not the sole factor to be considered and that a balance must be
struck between all relevant considerations as prescribed by NWA in the assessment of any
application. The High Court, and later also the Appeal Court, found that the DWS and the Water
Tribunal misinterpreted the provisions of section 27(1) of the NWA by regarding the redress factor
(section 27(1)(b)) as a prerequisite for the approval of a licence application. The court confirmed
that “all relevant factors”; must be taken into account for the approval of a license, and although
the Department has a discretion, it is still required to maintain a balance in its decision making.
One of the key components of a Catchment Management Strategy (CMS) is the development of
its water allocation principles for the catchment, taking into account the NWA S27. The CMS for
the Berg-Olifants is still to be developed and will be completed once (or if) the CMA has been
established.
Although the NWA, NWRS and potentially CMS provides a strong set of principles and
guidelines to guide the allocation process, a key finding of this study has revealed that, in
practice, these principles are not influential in the allocation process. Given allocation is
managed with a licensing process, DWS (or the responsible authority) must authorise a
water use if it meets the licensing requirements. DWS has no reason (and no mandate) to
withhold an authorisation on the hope (or even knowledge) that a more socially or
economically productive use of the perhaps constrained water resource is applied for at a
later date.
7 Available at https://cer.org.za/wp-content/uploads/2012/12/Goede-Wellington-SCA-2012-205.pdf
accessed 1 September 2017
26
Box 1 Validation and Verification Process and Compulsory Licensing
The DWS is in the process of validation and verification of water users in the Berg-Olifants Water
Management Area to ensure that all available water is accounted for. According to Regulation
1352, a person who uses water as contemplated in Section 21 of the NWA must register the water
use when called upon by the responsible authority to do so. The registration process requires that
all water uses, regardless of the pre-existing legal status thereof (i.e. already licensed), must
register. Water users in the Berg-Olifants WMA were asked to register their water usage, during
the qualifying period. This usage is then being reviewed by the DWS through the process of
validation and verification. Validation aims to confirm of the extent of water use (who, where, how
much & what for?) that took place during the qualifying period, to determine the extent of the
present water use, and thirdly, to determine the lawfulness of the uses. Therefore, registering
your water use does not guarantee that it is accurate or lawful. Verification refers to a process
that determines whether the water use identified in the validation process was lawful - it confirms
whether any previous laws would have authorised the use or limited the use in the Qualifying
Period.
Once the validation and verification process is completed, DWS may initiate the compulsory
licensing process which will enable them to reallocate water. Reallocation is a likely scenario for
the Berg WMA: The on-going Validation and Verification (V&V) process by DWS hopes to
discover that there is additional water to allocate as they suspect that registered water use is
higher than actual water use. Furthermore, once the V&V process is completed, and a sunset
clause is removed from the National Water Act that provides farmers with automatic riparian
rights, DWS may institute compulsory licensing in order to reallocate water across the region for
equity purposes.
4.3.1.3. Water Services Authority Supply
In the same manner as individual applicants, Water Service Authorities (WSA), which are almost
always municipalities (or water boards in other parts of the country), must apply for abstraction of
raw water from a water resource from the CMA/DWS. In turn, the WSA will have an agreement
with a Water Services Provider (WSP), often simply another department at the municipality, who
will treat raw water and distribute potable water to individual applicants under a service level
agreement (Figure 10). Roles and responsibilities of WSA and WSP are further outlined in Box 2.
However, water resources such as seawater (via desalination), and use of treated effluent, are
not directly allocated by DWS given its mandate to manage fresh water. The release of water to
a watercourse, from a wastewater treatment works, is licensed by DWS as one of 12 types of
uses of water that are regulated (along with taking from a water resource). But the allocation of
the treated effluent resource (who it goes to and in what quantity) is not. In the case of
desalination, DWS may only be involved in terms requests for infrastructure funding, approval of
any discharge, but not the quantification and allocation of the available resource. So, in addition
to the water resources managed on a regional scale by the CMA/DWS, the WSA may also have
access to water resources that are not allocated by DWS.
Applications by the WSA to DWS/CMA for additional water allocations are considered against the
same section 27 criteria, plus the manner in which the WSA has managed any existing water
allocations. Applications by a WSA for water from DWS/CMA have to be in alignment with the
27
recommendations and forecasts in the All Towns Reconciliation Strategy Study8. However, if
plans have been made for the development of a town/area above and beyond what is reflected
in the All Towns Strategy, then the WSA will struggle to gain approval for increased allocation
(Thompson et al, 2015).
The individual water users that require water services will include household users and larger
industrial users within the municipality, who for whatever reason, prefer to use municipal supply
water rather than abstract raw water. (Most users are connected to the municipal supply in
Saldanha Bay, due to the lack of local surface water resources.) Larger users, particularly
industrial or commercial users on greenfield or unserviced sites, will need to apply to the WSA for
access to water services. The basis on which this access is granted is determined by the Water
Services Act (RSA, 1997, chapter 1, section 8), which largely includes the practical considerations
of supplying the water, along with the “socio-economic and conservation benefits that may be
achieved by providing the water services in question”.
GreenCape’s scoping study (2015) highlighted that connections to water services are granted on
a first-come-first-served basis by the SBLM (and in essence, so are WULs if all necessary
requirements are met in the application, section 4.3.1.2). Although both SBLM and DWS
acknowledge that this is not an ideal situation, there are few alternatives in absence of better
information.
4.3.2. Development planning process
4.3.2.1. The IDP and SDF
Figure 11 illustrates the development planning process. The primary planning processes occur at
the local municipal level, through the Integrated Development Plan (IDP) (Figure 12). The IDP is,
simply put, a coordination and amalgamation of sector development plans which must reflect:
a) the municipal council’s vision for the long-term development of the municipality with
special emphasis on the municipality’s most critical development and internal
transformation needs;
b) an assessment of the existing level of development in the municipality, which must
include an identification of communities which do not have access to basic municipal
services;
c) the council and private sector’s development priorities and objectives for the council’s
elected term, including its local economic development aims and its internal
transformation needs;
d) the council’s development strategies which must be aligned with any national or provincial
sectoral plans and planning requirements binding on the municipality in terms of
legislation;
e) a spatial development framework which must include the provision of basic guidelines for
a land use management system for the municipality
8 The All Towns Strategy Study is a nationwide DWS project to forecast water requirements and supply in
all South African towns and villages (outside of the major water supply systems which have more in depth
reconciliation studies), and make recommendations on reconciliation of demand and supply, to ensure
sufficient water supplies into the future.
28
However, the overall development planning process is structured in a tiered manner which reflects
the different government spheres (see Figure 11 and Figure 12). In this regard, a local municipality
(LM) IDP has several planning processes above it that it is required to take into account. From
the top (in terms of hierarchy and spatial area of interest), the National Development Plan (NDP),
established by the national planning committee, has documented Vision 2030 that is intended to
shape all development in the country with the central aim to eliminate poverty and reduce
inequality by 2030 (NPC, 2011). This includes effective management of water resources and
services to support and promote a strong economy and healthy environment, effective water
planning that cuts across different economic sectors and spheres of government and reliable
water supplies for urban and industrial centres with increasingly efficient agricultural water use
(NPC, 2011).
The aims and vision of the NDP are implemented through a 5-yearly plan developed by Cabinet:
the Medium Term Strategic Framework (MTSF). The MTSF articulates Government’s
commitment to implementing the NDP and delivering on its electoral mandate as well as its
constitutional and statutory obligations (RSA, 2014). The priorities identified in the latest MTSF
(2014 – 2019) are being incorporated into the plans and programmes of national and provincial
departments, municipalities and public entities.
The NDP incorporates a National Infrastructure Development Plan 2012, with a series of
ambitious and far-reaching initiatives envisaged to transform South Africa’s economic landscape,
to virtually eliminate unemployment and improve the delivery of basic services. This plan listed
18 Strategic Integrated Projects (SIPs) which have been developed and approved by Cabinet and
the Presidential Infrastructure Coordinating Commission (PICC) to support economic
development. The identified projects will provide new infrastructure, assist in terms of
rehabilitating and upgrading existing infrastructure and will also play a crucial role in facilitating
the regional integration for African co-operation and economic development on the African
continent. This plan has relevance in areas where a SIP is located, of which Saldanha is one
(SIP5: Saldanha-Northern Cape corridor development, strengthening marine support capacity for
oil and gas and through the expansion of iron ore mining production, integrated rail and port
expansion).
At provincial level, Provincial Government develops a Provincial Strategic Framework that guides
all activity by Province, all focus areas and all projects completed. The local level IDP needs to fit
into, and correlate with, all of these broader-scale plans “above” it in the hierarchy presented in
Figure 11. A LM IDP includes all the projects/interventions which the LM plans to undertake. The
IDP can list private developments although it seldom does so for future development projects.
Furthermore, an IDP is also completed at district municipality (DM) level. According to the
Municipal Structures Amendment Act (RSA, 2003), a DM is responsible for developing an IDP for
the DM as a whole, including a framework for IDPs of all local municipalities in the area of the
DM. The DMs IDP is primarily aimed at coordinating development across the borders of the LMs
within its jurisdiction, ensuring alignment with provincial and national strategic plans and
supporting LMs where capacity is weak. Whilst this tiered approach does well to guide the
development planning process on a variety of spatial scales, a challenge exists in dealing
with the slow trickle down of information from higher level planning processes to the LM
IDP. Processes need to be aligned more timeously.
29
The Constitution stipulates that local government is responsible for the provision of water supply
and sanitation services, hence to operate as the water services authority and water services
provider unless this is outsourced under contract to a water board (RSA, 1996). To this end, the
IDP is the primary planning tool within which any water infrastructure interventions would have to
be outlined. The budget planning for the LM and the IDP go hand-in-hand i.e. projects within the
IDP are voted for at council level, and the required budget is set aside. Typically a 12-month
process is required for the completion of an IDP (see Table 5), and the required content is set out
in the Municipal Systems Act (RSA, 2000).
The level of detail contained within development plans differs at the different levels. Local-scale
planning is more detail specific whilst national scale planning has a greater strategic focus,
although will name individual projects where these are of national significance. For example, the
national spatial development framework includes mentioning of Saldanha Port because it is of
national importance.
The Spatial Development Framework is a sector plan that is regarded as the key input to the IDP,
and often elevated to the same importance as an IDP (demonstrated in Figure 11 and Figure 12).
The Provincial Spatial Development Framework (PSDF; Figure 11 and Figure 12) aims to
spatialize core projects, promote biodiversity, guide land use, guide bulk infrastructure
development, serve as a manual for municipal planning, create integrated land management
areas, facilitate cross-boundary collaboration, guide sustainable development initiatives, guide
allocation of government funds and optimise benefits of private sector investment.9 This is in
contrast with a more detail orientated LM SDF. The PSDF is guided by the Provincial Growth and
Development Strategy (PGDS, Figure 12).
The LM Spatial Development Framework (SDF) is a development plan focused on the spatial
planning in the LM’s area of jurisdiction. It identifies land suitable for future development projects
that are described within the IDP, as well as identifying changes on the urban edge. It also gives
a localised spatial dimension to development principles, objectives and projects and forms the
basis for the local government’s land use management system (DPLG, 2000). The development
of an SDF is regulated through the Spatial Planning and Land-use Management Act (SPLUMA)
(RSA, 2013). SDFs are completed at LM, DM, provincial and national level. Furthermore, the
Western Cape Provincial Government has recently initiated regional SDFs, at levels smaller than
provincial scale.
The National Environmental Management Act, Act 107 of 1998 (NEMA) allows for the
development of a national environmental management framework. In the same way as multi-
scaled SDFs, this has a provincial and local government equivalent (see Figure 11 and Figure
12). The regional environmental management framework is then incorporated into the regional
SDF. The process is important as it a key link between (environmental, spatial, integrated)
development planning, and whether a private sector project initiates: the project will only be
awarded an EIA if it is aligned with the various environmental management plans. The plans will
also dictate the content of the EIA.
9 See video on PSDF at http://northerncapepsdf.co.za/
30
Figure 11: A flowchart of the development planning process based on assessment of
legislation and discussion with stakeholders
31
Figure 12: Structure of key plans working within government (adapted from DRDLR,
2012)
Table 4: Generic IDP process (adapted from DRDLR, 2012). The dark green block
indicates the point at which sector development plans are incorporated into the IDP
IDP Process Months
1 2 3 4 5 6 7 8 9 10 11 12
Preparation
Analysis
Strategies
Project
Integration
Approval
Implementation
32
4.3.2.2. Water (and other inputs) in the IDP process
A fundamental feature of the IDP process is the integration of all sector plans into the IDP
document (dark green box Table 5). This includes the integration of the Water Services
Development Plan (WSDP) as the key plan for water services in the municipality. The compilation
of Water Services Development Plans (WSDP), by Water Services Authorities, is a planning
requirement laid out in the Water Services Act (RSA, 1997). In the event of the Water Service
Authority covering more than one local municipality, each local municipality should be
represented in the planning team for the Water Services Development Plan to ensure mutual
alignment of the WSDP and the water-related projects of the IDP (Haigh et al, 2008).
Sectoral departments/agencies are responsible for checking the alignment of the IDP with the
relevant sector’s plans and priorities during its drafting, while the national sector agencies should
provide a framework for the development of the sector plans. Insights from many officials
suggest that this often functions purely as a legislative requirement and there is
insufficient time to ensure meaningful commentary and input is given. One challenge,
highlighted by government officials during the 2016 workshop, relates to the different
timescales of the IDP and WSDP. An IDP only functions in five-year intervals whereas a
WSDP is generally in place for 20-30 year intervals (with an implementation plan for the
coming five years). An alignment of these two planning processes is challenging and
officials generally call for a repetition of WSDP information in the IDPs.
More significantly, the WSDP relates mostly to infrastructure and water services needs. They do
not, generally, include detailed consideration of future water resource or supply interventions.
These considerations are sometimes addressed in other planning documents of the LM or DM
such as water master plans, or only addressed by DWS, for example in the All Towns Strategy
Study. There is no explicit requirement for the reconciliation interventions listed in the All Towns
Strategy Study to be included in the WSDP, nor in the IDP. Typically the IDP will simply include
a recognition of currently available water resources, and potential future water availability, by
repeating information that may be available in the WSDP or other documents. This demonstrates
the challenge originally noticed; that water and economics are treated as independent variables
to be taken into account in the planning of the other, rather than recognising the two are
intrinsically linked (section 1.2). During the 2016 workshop, it was proposed that the Catchment
Management Strategy (CMS) be included as one of the sector plans input to the IDP since water
resource to support future water supply interventions are not currently adequately considered in
the WSDP. However, the CMS does not address local supply augmentation problems, so this
remains a problem.
The Local Economic Development (LED) plan is another sector plan required as input into the
IDP process and it is an approach towards economic development which allows and encourages
local people to work together to achieve sustainable economic growth and development thereby
bringing economic benefits and improved quality of life for all residents in a local municipal area.
The LED plan is intended to maximise the economic potential of all municipal localities throughout
the country and, to enhance the resilience of the macroeconomic growth through increased local
economic growth, employment creation and development initiatives within the context of
sustainable development. The “local” in economic development points to the fact that the political
jurisdiction at a local level is often the most appropriate place for economic intervention as it
carries alongside it the accountability and legitimacy of a democratically elected body. The LM is
33
a key player in shaping the local economy and should formulate policies and provide the
infrastructure that fosters economic growth (DPLG, 2000).
The private sector (from households to businesses) is also a key consideration within the LM IDP.
The private sector provides input to the IDP and SDF carried out at LM level through the
stakeholder participation processes. They are also subject to the output of these plans.
Development applications from the private sector to the LM that are not in line with the SDF and
IDP will not be approved or require additional levies for infrastructure development. Participants
at the 2016 workshop highlighted the need to promote increased communication, integration and
coordination between the private sector and local government, specifically because the private
sector felt it had significant insight to private developments that would impact the area. The current
status quo is that the private sector is generally unwilling to disclose specific development plans,
to protect market competitiveness, and also due to related impacts on land use and property
prices. The LM is only consulted once permissions or expansion authorisation are needed.
This situation is sub-optimal as if the LM is aware of future industrial development plans,
it can factor in municipal services and associated projects accordingly, and industry could
play a key role in economic and social development. If better transparency could be promoted,
it was felt that the LM could put together a more informed IDP that is considerate of industry’s
needs. This broadly demonstrates a recurring issue with the IDPs, specifically that the
depth of commentary given during the participation and input process, is typically rushed
and more legally obligatory, rather than promoting detailed, useful contributions. This
problem has been stressed an issue both internally (i.e. from governmental departments) and
externally through the private sector.
In summary, development planning currently functions in a top-down manner. Overarching
development goals are set at national level and inform planning at lower levels but are not
necessarily compatible with LMs’ available resources, budgets and capacity. The IDP, which at a
LM level is generally seen as the primary planning tool of a local municipality, is not timeously
informed by higher level documents and does not feedback to them. Contributions to the IDPs
are also often rushed and serve to tick a legislative requirement but lack meaningful contributions
from both government and the private sector. The IDP is also the touch point between
development and water resources planning, through the integration of the WSDP as a sector plan,
but does not adequately consider future water resource availability and the link to development.
4.3.3. Water Services and Resources Development Process
4.3.3.1. Water Resource Development
A review of the processes involved in water supply augmentation decisions, at both a resources
and services level is necessary for a complete view of economic development and water
resources planning. Figure 13 depicts a summary of the water resources and services
development process.
Roles and responsibilities of DWS and WSAs
A Water Services Authority (WSA) is responsible for all planning necessary to ensure sufficient
water is available to users i.e. all elements of water resources planning for municipal supply (Box
2), with support from DWS. In the case of regional schemes that supply more than one WSA
(which were historically generally established by DWS), DWS is responsible for oversight and
34
coordination of this planning, however relies on the input from WSAs to coordinate requirements
between the various users of the regional schemes.
Soon after the establishment of the NWA (1998), DWS begun the process of quantification of
available water resources, first with the Internal Strategic Perspectives (i.e. DWAF, 2004), then
with the Water Availability Assessment Studies (i.e. DWAF, 2008), both of which were projects
completed across the country separated into individual projects per water management area
(some were grouped). In parallel, in the mid-2000s DWS’s Directorate of National Water
Resources Planning was aware that the water requirements in the major metropolitan areas and
major schemes were soon to outstrip availability and that the country’s development priorities
necessitated increases in supply in most areas. In response, the DWS initiated the reconciliation
strategies for all the major schemes in order to guide the necessary water resources planning (i.e.
DWAF, 2007), although it is ultimately the responsibility of the WSAs using the major schemes to
ensure supply. In a reconciliation strategy, available water supplies are compared to current and
projected future water requirements. Where there are shortfalls between future water
requirements and currently available supplies, water resource interventions are proposed (along
with necessary actions i.e. presented as a strategy), to reconcile supply and demand, such as
demand reduction and increasing availability.
The DWS then initiated the ‘All Towns Reconciliation Strategy study’ which included the
development of water reconciliation strategies for towns, villages and clusters of villages (i.e. all
population that is or should be served by a WSA) in 2008, concluding the first phase in 2011. The
project was intended to provide DWS with oversight of and guidance for the LMs in their planning
of water resources interventions. The DWS has therefore completed reconciliation strategies at a
local, regional and national level. These strategies vary in complexity, depending on the
geographic focus and complexity of the water supply scheme being assessed.
The gaps between DWS and WSAs
Whilst it is the responsibility of the WSA to provide water resources, the DWS has (as outlined
above) acted in a guiding and advisory capacity in the generation of reconciliation strategies, for
regional schemes and local water resources. However in some cases, WSAs continue to look to
DWS for water resources planning support, focussing their sights on services provision only. The
following challenges are listed in the NWRS2 with respect to managing regional water
infrastructure and supporting local government in the delivery of water services:
“Weak performance in the management of water supply and sanitation services by many
municipalities, which compromises services;
Unclear responsibilities for water resources development at the local and regional level, and
for regional bulk services outside of the existing water board service areas; and
Governance and performance-related problems within some of the existing water boards”
(DWA, 2013a, pg. 59).
In response to these challenges, Regional Water Utilities have been proposed and will be created
from the 12 existing water boards (further described below), to manage regional water resources
and regional bulk water and wastewater infrastructure in terms of a mandate from the DWS.(DWA,
2013a; DWA, 2014). However, to date, little progress has been made with regards to the
establishment of a Regional Water Utility in the Berg WMA.
35
Box 2: Responsibilities of Water Services Authority and Water Services Provider (quoted
from Riemann et al, 2011, pg. 9)
Financial support for water resources development at WSAs
In order to coordinate water resources investment requirements from WSAs, the DWS proposed
the initiation of a National Water Sector Infrastructure Investment Framework; part of the National
Water Investment Framework (DWS, 2013a). The National Water Investment Framework was
intended to document the costs and financing options of the entire water sector from source to
tap to waste and back to the source, and include the investment requirements of DWS, CMAs,
water boards, WSAs and WSPs (DWA, 2013a). Infrastructure requirements proven in the All
Towns Reconciliation Strategies, and the recommendations in the reconciliation strategies for the
larger schemes, were to be included in the Infrastructure Investment Framework (DWA, 2013a).
A presentation to Parliament in 2015 provided an update on progress made by DWS on the
investment framework and included costs for the national water resource development and
municipal water services development (DWS, 2015c). This update revealed that funding had been
increased by R13 billion/year but that there was still a significant funding shortfall for water and
sanitation infrastructure of R35 billion/year. The National Water Investment Framework was
intended to inform a Water Investment Strategy, followed by a Water Investment Plan (DWS,
2015c). The current (2017) status of the investment framework is not known.
Duties of Water Services Authorities
Water Services Authorities have the following primary responsibilities:
Realisation of the right to access to basic water services: ensuring progressive realisation of the right to basic water services, subject to available resources (that is, extension of services), the provision of effective and efficient ongoing services (through performance management, by-laws) and sustainability (through financial planning, tariffs, service level choices, environmental monitoring).
Planning: preparing water services development plans (integrated financial, institutional, social, technical and environmental planning) to progressively ensure efficient, affordable, economical and sustainable access to water.
Selection of water services providers: selection, procurement and contracting water
services providers (including itself).
Regulation: of water service provision and water services providers (by-laws, contract
regulation, monitoring, performance management).
Communication: consumer education and communication (health and hygiene promotion, water conservation and demand management, information sharing, communication and consumer charters).
Duties of Water Services Providers
The main duty of water services providers is to provide water services in accordance with the Constitution, the Water Services Act and the by-laws of the water services authority, and in terms of any specific conditions set by the water services authority in a contract. A water services provider must publish a consumer charter which:
is consistent with by-laws and other regulations;
is approved by the water services authority; and
includes the duties and responsibilities of both the WSP and the consumer, including conditions of supply of water services and payment conditions.
36
The DWS has recently (mid-2017) launched the National Water and Sanitation Master Plan
(NWSMP), whose aims are similar to the National Water Investment Framework, and presumably
are replacing it. The NWSMP cites un-coordinated planning and delays in infrastructure
development and inadequate financing of water resources as water sector challenges to which
the NWSMP plans to respond (among others). The NWSMP intends to guide, integrate, and
facilitate:
Infrastructure development, refurbishment, operation & maintenance, fighting Non-Revenue
Water;
Institutional arrangements & roles/ responsibilities; and
Informs prioritisation or criteria of infrastructure development (amongst others). (DWS
2017b)
The NWSMP is planned for completion by mid-2018. In the immediate absence of the Water
Master Plan, the process by which water resource infrastructure requirements are prioritised by
the Minister is not clear. However, once the infrastructure need has been established, the Minister
of Water and Sanitation determines whether the project is classified as social and economic
stimulus infrastructure or commercial infrastructure, or both. Social infrastructure supplies the
basic water requirements of municipal water users in rural areas and economic stimulus
infrastructure promotes economic development where there are no current users or where the
users can’t afford the supply (DWS, 2015b). Both are hereafter referred to as social infrastructure.
Commercial infrastructure is seen as commercially viable infrastructure. The distinction of the
project or infrastructure as social or commercial is at the discretion of the Minister. Typically, social
infrastructure is funded on-budget from the fiscus with charges being set to recover operational
and nominal asset costs, while the latter is funded using commercial off-budget finance with
charges being set to recover the full financial cost of operation and debt repayment (Pegasys,
2012). There are cases where infrastructure is determined to be a mix of the two, with the
infrastructure being funded on-budget but where commercial users are charged full costs or where
the social component within commercial projects is funded by the State (DWS, 2015b). Feedback
from the 2016 workshop (Appendix 1) highlighted that access to funding has proved complicated
for projects which comprise both social and commercial classification. For example, the WCDM
application for funding to DWS for a desalination plant was deemed to be partially social and
therefore grant funding for part of the infrastructure investment was approved, with the remainder
of the funding (R200 million of R450 million) to be raised by the municipality (personal
communication, Minnaar, 2016). The municipality had not planned for such a large capital
investment (hence no capital investment tariffs were collected) and did not have the funding to
cover the commercial component of the plant and there was resistance from local industry who
saw this as too expensive. Finding funding for new infrastructure is frequently a challenge in
municipalities when considering that maintaining and rehabilitating existing infrastructure is
already under-funded. Access to funding is further complicated due to the capacity required
to raise off-budget finance often being greater than that which is available at the LM.
The entities responsible for the development of the infrastructure are split according to whether
the project is considered on or off-budget. The Water Trading Entity (WTE) functions within the
administration of DWS and is responsible for the development of new on-budget infrastructure,
amongst other water resources and infrastructure management responsibilities (DWA, 2013b).
However, it has been acknowledged that the WTE is not the most appropriate or efficient
institutional arrangement for managing DWS’s water infrastructure and an alternative model has
been proposed, with a dedicated organisation focused on national water resources infrastructure
37
(DWA, 2013a). It is not clear if the institutional framework for this proposed new agency has
progressed or has been postponed (personal communication, Bosman, 2016).
The Trans-Caledon Tunnel Authority (TCTA) is a public entity responsible for the development
off-budget bulk raw water infrastructure development, specializing in project finance,
implementation and liability management (DWA, 2013b). TCTA is mandated to raise funds from
commercial domestic and international funders. TCTA generally develops infrastructure that is
commercially viable and the full cost of the investment is recouped through the revenue from
water sales over a twenty-year period. The Berg Water Project (the Berg River Dam) was funded
off-budget by TCTA and financing raised from ABSA Bank, Development Bank of South Africa
and the European Investment Bank (Pegasys, 2012).
Additionally, the new raw water pricing strategy (DWS, 2015b) proposes a change to the manner
in which revenue for on-budget infrastructure is generated – from a Return-On-Assets (ROA)
approach to a Future Infrastructure Build Charge (FIBC). The FIBC will be a determined on a
national basis (DWS, 2015b) as opposed to the scheme-specific ROA. The FIBC will support the
development of social infrastructure and “provide for the costs of investigation, planning, design,
construction and pre-financing of new infrastructure and the betterment of already existing
infrastructure” (DWS, 2015b:19). FIBC will be paid by municipal, industrial/mining and high
assurance use categories only. Agriculture’s infrastructure tariffs, are still capped.
The Capital Unit Charge (CUC) is used to generate revenue for commercial infrastructure and will
continue to operate according to the existing 2007 pricing strategy. The CUC is scheme-specific
and calculated on water used, not necessarily on water provided into the scheme. The CUC will
provide for the debt service requirements on the project, once the financial viability of the project
has been established with the water users. All water users supplied from the scheme, with the
exception of the social component, will be liable for the CUC (DWS, 2015a). The CUC ceases
once the debt has been repaid, and thereafter will attract FIBCs. The pricing strategy therefore
provides for the raising of revenue for the development of water resources through appropriate
raw water tariffs, however an existing challenge with the on-budget and off-budget schemes is
that there is a funding gap between the DWS funded on-budget developments and the larger,
TCTA financed off-budget development projects. The difficulty experienced by a WSA in raising
financing for the non-grant component of a water project (typically partially funded by RBIG) is a
“funding gap” i.e. there is funding available for local water resource development or infrastructure,
but it will likely need to be raised from commercial sources or development finance institutions. If
the WSA can’t raise this due to their lack of credit-worthiness or ability to raise tariffs, then the
project will not happen. This is the case in SBLM. To this end, Regional Water Utilities (RWUs)
have been proposed and will be created from the 12 existing water boards with the mandate to
fill this ‘gap’ (DWA, 2013a; DWA, 2014).
The mandate of RWUs will be expanded to include the development and management of regional
water resources, regional bulk water services and regional wastewater infrastructure. They will
be responsible for the financing, development, management, operation and maintenance of
regional bulk water infrastructure in an efficient and effective manner to meet the social and
economic development needs of current and future users to achieve the objectives of integrated
water resources management. The RWUs will also play a strong secondary role of supporting
municipalities by providing water services on their behalf to users or by providing services directly
to municipalities on a contractual basis, provided this does not detract from their ability to fulfil
38
their primary functions. In contracting with WSAs to provide affordable and sustainable water
services in accordance with S78 of the Municipal Systems Act (RSA, 1997), RWUs will be able
to complement local government capacity through the benefits of economies of scale by
integrating risk management and by leveraging finance for commercial water supply projects
(DWA, 2014).
39
Figure 13: A flowchart of the water resources and services development process based
on assessment of legislation and discussion with stakeholders
40
4.3.3.2. Water Services Development
Water services are developed by WSAs, typically municipalities. Their legislative mandate is the
Water Services Act (RSA, 1997). The WSAs are responsible for preparing a Water Services
Development Plan (WSDP) and ensuring its implementation. The Service Delivery Budget
Implementation Plan (SDBIP) ensures that funding from the IDP is channelled towards the
implementation of the WSDP (illustrated in Figure 13). The WSDP must provide information on
the “future provision of water services and water for industrial use” including details on “the
proposed infrastructure necessary”, “the water sources to be used and the quantity of water to be
obtained from and discharged into each source” and “the estimated capital and operating costs
of those water services and the financial arrangements for funding those water services, including
the tariff structures” (Water Services Act, RSA, 1997). The WSDP must therefore have a good
degree of alignment with the regional or bulk water resources development planning process,
where relevant, and catchment management strategies. As per the NWRS2 (DWS, 2013a), it is
the responsibility of local government, to align their functions to water resource management
functions and institutions. This includes monitoring all relevant plans to inform their planning, and
the alignment of this planning with the availability of water resources, water supplies and bulk or
regional infrastructure plans. A sub-section of the WSDP; the Water and Sanitation Master Plan
plays a key role in determining the feasibility of water services infrastructure.
Alignment between DWS and the WSA becomes critical when financing of the water services
infrastructure is to be approved as part of a Council Budget adoption. The WSDP is a sector report
included in the IDP and approval of the WSDP plans are reliant on their inclusion in the IDP. Grant
funding from the state, as noted above, is only relevant for infrastructure that fulfils a social
objective. These grants may be sourced from the Treasury’s equitable share (for operating
grants) or Municipal Infrastructure Grant (MIG) for capital funding (amongst other capital
grant mechanisms). However, if the infrastructure is not classified as social and does not
qualify for grant funding, the municipality will need to generate revenue for the
infrastructure through user tariffs. These could be through user charges, rates levies,
capital funds or loans. The development of local sources of water (groundwater, reclaimed
water or desalinated water) may need to be partially financed by the municipality where a
component of the infrastructure is deemed to be commercial.
4.4. Key forums for integrated water resources and economic development
4.4.1. Overview
One key outcome from the local prioritisation tool (chapter 7), is that it brought government
officials from a variety of departments within the SBLM together to discuss and consider which
key variables need to be considered (at any given time) within the MCDA tool. The tool was
developed with the target of implementation within a single municipality, and the tool promotes
cooperative governance and integrated planning.
A ‘regional tool’ has been created with the aim of providing the information required to make
integrated water resources and development planning decisions (chapter 6). The ‘tool’ is a series
of data and maps which allows planners to forecast different scenarios and make decisions
through a comparative analysis of the outcomes. The development of the tool was guided by
41
project findings and stakeholder engagement as the project progressed. The emphasis in the
investigation of forums, was to determine where, how and by whom, this tool could be used to
inform greater integrated planning. A key finding from the analysis of legislation, and the
enquiry of perceptions, is that the DWS (or any DWS-led platforms) is not the appropriate
body to lead the kind of integrated planning being recommended by this project (section
4.5). Continued insights suggested that provincial government have a key role to play in
supporting local municipalities and are the most appropriate house for strategic oversight of water
resources and development planning (section 4.5). A focus was therefore placed on forums that
have diverse, well-represented stakeholders, meet regularly, include water resource and
development planning (at least each in part) and which are preferably chaired by the provincial
government. Whereas findings in section 4.3 are largely relevant nationally, with local examples,
the assessment of the forums operating for integrated water resources and development planning
is based on local (Berg WMA) information, with local relevance.
Table 5 contains details of forums identified as most relevant for integrated water resources and
economic development planning (and therefore the potential implementation of tools/approaches
developed in this project) in the Berg WMA.
In addition to those listed in Table 5, the steering committee for the current DWS project for the
Classification of Water Resource and establishment of Resource Quality Objectives (RQOs) for
the Berg River was established in November 2016. The steering committee consists of a number
of role-players from various provincial and local government organisations (including, but not
limited to, representatives from DWS, DRDLR, DEADP, DOA, CCT and numerous district and
local municipalities), conservation authorities (Cape Nature) and non-governmental
organisations. It is anticipated that the project will be completed towards the end of 2018 at which
point the established (and gazetted) RQOs and Management Classes will be incorporated into
the CMS for the Berg-Olifants CMA.
Given the overall aim and intention of the Water Resource Classification System (WRCS) process
(Dollar et al, 2010, Box 3), the current DWS project in the Berg, and the steering committee for
the project, has the potential to provide a vehicle and platform for integrated water resources
planning and economic development. The WRCS process intends to determine the intended
development trajectory and corresponding condition of water resources, based on the economic,
social and environmental cost/ benefit of the various development and conservation scenarios.
The process requires that trade-offs between development (increased use of water resources)
and condition of water resources (ecological functioning) be quantified. The binding nature of the
RQOs therefore has potential implications for development in the area in terms of both (a) access
to water resources and (b) environmental impacts on the water resources.
However, the process is likely to lack the granularity in detail to fully translate to tangible (go / no
go) impacts on particular developments. The process is also driven from DWS and set according
to the gazetted process for establishing the Management Class and RQOs, with relatively little
potential for participatory engagement. The RQOs will likely only impact on the targets particular
developments will need to meet in their effluent discharges, for example. Nevertheless, increased
involvement from DEADP (the spatial planning department and environmental authorisation) is
required, in order fully understand potential development implications and provide feedback on
where development is already planned.
42
The analysis has led to a recommendation of the provincial government’s Sustainable Water
Management Plan (SWMP) as the “best fit” existing forum to implement the regional tool, and
adopt the integrated planning promoted here. Hence, additional detail is given on the SWMP in
section 4.4.2.
43
Table 5: Characteristics of key water and planning forums in the Berg
Forum Convening
Institution Stakeholders Principle Focus Areas and Mandate Frequency Other Comments
Berg River
Partnership
(BRP)
DWS
DWS, DEADP, BRMIB, CWDM, DLG,
Drakenstein Municipality, Stellenbosch
Municipality, WCDM, Witzenberg
Municipality, Cape Nature, Stellenbosch
University, UCT, UWC, GreenCape,
Casidra, WWF, Living Lands, Fruit Look,
Private Land-owners
Predominantly focused on water quality and
ecological restoration. The forum also
presumably serves as a stakeholder coordinating
forum.
Quarterly
Last met in October 2015;
forum has been reviewed in
latter half of 2017
Berg River
Improvement
Plan
DEADP DEADP, DWS, DEDAT, Cape Nature,
DoA, DLG
Focused on establishing a water stewardship
programme for the Berg River with emphasis on
ecological restoration, water quality and
agriculture
Quarterly
Meetings
Similar forum currently being
created for the Breede River.
“BRIP 2”
Western
Cape
Sustainable
Water
Management
Plan (SWMP)
DEADP
DEADP, DWS, DEDAT, GreenCape,
Municipalities, DOA, Irrigation Boards,
Isidima
Broad mandate which includes most issues
relating to water – conservation, management,
education, info sharing etc.
Quarterly
Meetings
Consultants appointed by
DEADP to review mandate and
goals
WC/WSS
Steering
Committee
DWS
DWS (head and regional offices), DoA,
CoCT, WSPs and WUAs. District and
Local Municipalities, BGCMA
Coordination of the implementation of the water
resources interventions within the reconciliation
strategy for WCWSS
Bi-annual
Meetings
Last met in March 2016 and no
longer meeting; contract for the
support for the continuation of
the strategy has lapsed.
Premiers
Coordinating
Forum
Western
Cape
Premier’s
Office
Premier, MECs, Director-General, Heads
of Department, Mayors and municipal
Managers
Promoting cooperative governance between
provincial government and municipalities to
ensure integrated, effective and efficient service
delivery
Quarterly
High level but also deals with a
range of topics, typically those
which are most pressing and
topical
Provincial
Liaison
Committee
DWS Provincial heads of department from
DEADP and regional heads from DWS
Focus on cooperative governance, one
environmental policy and conflict resolution Quarterly
Seems to have gone stagnant
and has not met recently
Municipal
Planning
Heads Forum
DEADP DEADP, DOA, DRDLR, SALGA Principal focus is on legislative and land-use
planning issues relating to the LUPA. Quarterly
Water not currently a key
discussion point in this forum
44
Box 3: Water Resource Classification System and Resource Quality Objectives for the
Berg River
The NWA sets out to ensure that the nation’s water resources are protected, used, developed,
conserved, managed and controlled, in a manner that promotes equitability, efficiency and
sustainability for present and future generations. To this end, the Act prescribes a series of
measures which are intended to ensure comprehensive protection of water resources so that
they can be used sustainably. These measures are to be developed progressively within the
context of the NWRS2 and catchment management strategies (DWS, 2017a). In particular,
Chapter 3 of the NWA provides for:
1. The development of a Water Resources Classification System (the WRCS)
2. The setting of a Management Class and Resource Quality Objectives (RQOs)
3. Determination of the Reserve
Water resources are categorized according to specific classes that represent a management
vision of a particular catchment. This is done analysing the current state of the water resource
and defining the ecological, social and economic aspects that are dependent on the resource.
The implementation of the WRCS therefore requires an assessment of the costs and benefits
associated with utilization versus protection of a water resource. The WRCS defines three water
resource classes, reflecting a gradual shift from resources that will be minimally used, to
resources that are heavily used. Water resources must be classified into one of the following
classes:
Class I water resource: is one which is minimally used and the overall ecological condition of
that water resource is minimally altered from its pre-development condition.
Class II water resource: is one which is moderately used and the overall ecological condition of
that water resource is moderately altered from its pre-development condition.
Class III water resource: is one which is heavily used and the overall ecological condition of that
water resource is significantly altered from its pre-development condition
The classification of water resources represents the first stage in the protection process and will
result in the determination of the quantity and quality of water required for ecosystem
functioning (the Reserve) as well as maintaining economic activity that relies on a particular
water resource.
The RQOs capture the established Management Class of the classification system and the
ecological needs of the Reserve, representing this as measurable management goals that give
direction to resource managers as to how the resource needs to be managed (DWS, 2017a).
RQOs describe, among other things, the quantity, pattern and timing of instream flow; water
quality; the character and condition of riparian habitat, and the characteristics and condition of
the aquatic biota. These RQOs are essentially narrative and qualitative and are aligned with the
vision for the resource. Because RQOs are descriptive, and generally easy to understand, they
are also meaningful to stakeholders, as well as the responsible managers, and give direction for
whatever action is necessary to achieve the vision for the resource. These RQOs are gazetted
and are thus supported by law.
45
4.4.2. Western Cape Sustainable Water Management Plan
The Sustainable Water Management Plan (SWMP) for the Western Cape Province was
developed following the recommendations made at the DWS-coordinated Water Indaba held in
Cape Town during November 2009. Its development was undertaken collaboratively by the
Western Cape Government and the then National Department of Water Affairs: Bellville Regional
Office. The SWMP recommended short (1-5 years), medium (6-15 years) and long-term (+16
years) actions and projects towards achieving integrated and sustainable management of water
in the Western Cape (DEADP, 2012).
One of the actions in the original plan, was the establishment of a steering committee to oversee
the continued implementation of these actions and projects, beyond the initial development of the
plan. The overall aim of the SWMP is to guide sustainable water management in the Province
towards meeting the growth and development needs of the region, without compromising
ecological integrity. Four strategic goals form the cornerstone of the SWMP, and therefore also
guide the mandate of the steering committee (Table 6). Goals 1 and 2 are of key interest in the
context of this project, as they relate directly to the key challenges that this project is seeking to
address.
Table 6: Strategic Goals of the SWMP
Goal Description
Goal 1 Ensure effective co-operative governance and institutional planning for sustainable water management
Goal 2 Ensure the sustainability of water resources for growth and development
Goal 3 Ensure the integrity and sustainability of socio-ecological systems
Goal 4 Ensure effective and appropriate information management reporting and awareness-raising of sustainable water management
At the start of 2017, the SWMP commenced with a strategic review process. This is a requirement
of the SWMP – specifically that the relevance of its mandate and strategic goals are reviewed
every five years. The review, sub-contracted from DEADP to a consultant service provider, has
provided a unique opportunity to re-evaluate priority areas requiring renewed focus and an
opportunity for some of the key finding from this study to be incorporated into the review process.
Many of the findings in this project have been echoed during the review of the SWMP mandate.
Amongst the key insights in the SWMP review has been the recognition for the importance
yet inadequacies of the IDP in local water resources development planning, a recognition
of the potential role of provincial government in providing support to LMs, and recognition
of an urgent need to promote deeper integrated planning moving forwards.
During the SWMP review, a workshop was held in which stakeholders were asked to “vote” for
key goals and objectives requiring priority action moving forwards (i.e. to be incorporated into the
SWMP mandate). The overwhelming majority of stakeholders voted for objectives listed under
goal 1 – Cooperative governance and planning (Figure 14). Within this, the objective which
received the most votes was Objective 1.3 – Integrate sustainable water management with
planning and ecological sustainability.
46
Given the mandate of the SWMP (and particular the first two of its four strategic goals), its
wide range of stakeholders, its regular quarterly meetings and the initial outcomes from
the review process it is currently undergoing, it appears to be the most appropriate forum
to address the misalignment of planning approaches, for potential implementation of the
regional tool developed here, and more broadly, the forum to lead integrated water
resources and development planning in the Berg.
This recommendation was discussed with stakeholders at the third project workshop in June 2017
(Appendix 2). Members from the DEADP and stakeholders of the SWMP gave general support
for the presented conclusion: that the WCG’s SWMP forum is a good vehicle to take on the
integrated water and economic development coordination that is being recommended by the
project (further details in Appendix 2). It was noted that the SWMP has been lacking a focus on
the planning and development aspects of coordinated water resources planning. As this is within
WCG’s mandate, the involvement of the correct people and information from spatial, economic,
and development planning, with those involved in water, must be realised. However, it was noted
that the SWMP needs to improve working relations with other key role players such as the Local
Municipalities (LMs) and Department of Water and Sanitation (DWS), to ensure that the
recommendations for integrated water and economic development are translated into
authorisation decisions Indeed, the tools were seen as having the potential to play an important
role in facilitating greater integrated planning for the region.
Figure 14: Image of prioritised objectives (under “cooperative governance and planning”)
as voted for by stakeholders during the SWMP mandate review workshop (May 2017)
4.5. Insights
Currently, the economic impact of water resource decisions, and the water resource implications
of economic decisions, are not fully considered. Water is taken into account in economic planning
47
(and vice versa), but generally as independent variables rather than accommodating the co-
dependence of water and economics (chapter 1, chapter 3, and section 4.2). Optimally integrating
water resources and economic development planning would see
i) available resources allocated for optimal benefit, including consideration of economic
development and job creation;
ii) the economic benefit and water intensity of different water uses informing
development decisions;
iii) the ability of future development to fund water resources interventions taken into
account in the viability of interventions; and
iv) overall water resources challenges are taken into account in regional development
decisions.
Documenting the processes of water allocation, development planning and water services and
resources development, and discussing these with stakeholders (workshop 2), provided key
insights into water and development integration points, gaps and opportunities.
What is needed to realise integrated planning is largely provided for in legislation. In principle,
economic development is detailed as part of the preparation for an IDP, and can then be shared
with those preparing the WSDP such that future developments are catered for. The WSDP is
required to include necessary planning for future water resources. The WSDP is then reflected in
IDP and required interventions are budgeted. The provincial government oversees the
development of IDPs, and is responsible for ensuring cross-boundary integration. The CMS
outlines water resource availability in the region and should also be taken into account in the
WSDP and hence IDP process (and vice versa, the necessary development should inform
resource planning in the CMS). Furthermore, there is widespread recognition that water is a key
(and limiting) resource for economic development and, given this status, it poses an inherent risk
to the economic development agenda. It is also widely recognised that there are touch points,
particularly through local IDPs and regional SDFs, where the potential exists for greater integrated
planning that identifies imminent development plans and objectives and aligns these with
sustainable water resource planning.
However, there are significant shortcomings in the processes resulting in them falling short of
potential. Not all private developments are known during the development of an IDP, which can
lead to a situation where the LM is incapable of providing water resources to development. More
fundamentally, in most cases the WSDP is failing in its consideration of water resources and only
considers service requirements, thereby assuming an increase in the resource supplying that
service is feasible. The Berg-Olifants CMA is not established, hence there is no CMS to support
the IDP and WSDP process. Contributions to the IDPs, including provincial oversight, are also
often rushed and serve to tick a legislative requirement, but lack meaning from both government
and the private sector.
When this project commenced, the working assumption was that the point of impact, to
ensure more strategic use of water in terms of economic development, was via the water
allocation process itself. However, the analysis of the variables that are prioritised when it
comes to water allocation (the use of the word ‘prioritised’ in this context proved to be a
misnomer), and the insights from workshop 2, show that, notwithstanding some
predefined legal obligations, water allocation functions as a licensing or authorisation
process. If a water use licence application meets the necessary application requirements,
48
and there is water available to allocate, the application is passed. The DWS has no scope
(in their mandate) to, for example, wait for other potentially more economically beneficial
uses of the same (limited) water resource and allocate water to this later application.
Further, there is no scope for DWS to influence the kind of applications that will be
submitted (unless the NWA was amended). The DWS (or any DWS-led platform) is
therefore not the appropriate body to lead the kind of integrated planning being
recommended by this project.
On a local scale (and where water resources are local), it is the responsibility of a LM to ensure
economic planning in the IDP is in line with water resources planning contained in the WSDP
(given the role of a WSA). The reality however is that LMs are generally not capacitated to plan
accordingly. The WSAs are generally under-performing in terms of water resources planning, and
over-relying on DWS to provide this role, which it provides only for regional schemes. Where
adequate plans are made by WSAs, implementation slows due to hurdles such as financing. In
the Berg WMA, as with many other areas, water resources are shared between many LMs, DMs
and a metropolitan municipality. Because of the regional nature of the scheme, DWS therefore
does coordinate planning for water resources interventions to reconcile future demand on the
WCWSS. However, clearly this coordination is also insufficient at the level required, given
competing applications have been submitted from various WSAs using the WCWSS, which DWS
has been recommended to urgently address (DWS, 2015a).
The various assessments demonstrate that provincial government is the most appropriate body
to provide this strategic oversight and integration role. Provincial government has the potential to
identify and recognise threats to development plans as a consequence of (water) resource
constraints, and provide support to LMs - both in terms of appealing for funding from national
government for local water resource development and by providing capacity and support in driving
these processes (and in the necessary water resources assessment). Related to this, the analysis
led to a recommendation that the provincial government’s Sustainable Water Management Plan
(SWMP) is the “best fit” existing forum to promote the strategic planning promoted here. This
oversight role should not only focus on LMs requiring capacity support. The water resources
needs and development within CoCT must also be considered by the provincial government, in
terms of implications for the province. Economies of scale are not currently being adequately
considered between WSAs within the Berg to prioritise the most appropriate water resources
infrastructure, prior to the related WULAs being submitted to DWS for their consideration. For
example, perhaps a desalination plant in CoCT should be built larger to relieve more of CoCTs
demand on the Berg, enabling higher use from the WCWSS in WCDM.
The analysis of legislation, supported by stakeholder engagement, highlights there is no clear
mandate for the necessary support to local government to fulfil water resources development.
The ‘best fit’ may sit with the department of local government, and Provincial Treasury. It has
been suggested that this provincial oversight and support to LMs will be easier when SPLUMA
and LUPA are fully implemented. Fundamentally however, tension remains between DWS and
WCG over mandate in water resources, and is routinely evident in meetings or workshops with
the two present. Although there have been attempts to resolve this at a high level, tension
remains.
4.6. Way forward
These insights and recommendations are summarised as the way forward below:
49
The WSAs in the Berg WMA require support for water resources planning, and in the
absence of a Regional Water Utility or water board (both of which should be explored as
possible avenues to resolve this issue) this support is best provided by WCG. However,
tension remains with DWS regarding WCG’s involvement in water resources. A facilitated
discussion over mandate and the potential for the formation of a separate entity (such as a
water board) is recommended between WCG and DWS. It is recommended that the
discussion be held at the highest level, and is facilitated by experienced external strategic
planners.
In order for WSAs to perform better in terms of water resources planning, water resources
planning must be strengthened in the WSDP process, and as such, water resources
planning strengthened in IDP. It is recommended that provincial government, (through the
SWMP and other mechanisms), oversees the water resources planned at LM level in
WSDPs, coordinating these between neighbouring LMs / DMs.
A gap exists in the support to LMs in the preparation of project finance applications to DWS
(and other funding mechanisms, including commercial financiers), and WCG should explore
the options in terms of providing support to LMs for these applications and/or the
establishment of a regional water fund to assist in the financing of water resource
infrastructure.
It is recommended that WCG use the outputs of the regional tool to guide them in their
assistance to the LMs, in terms of appropriate development trajectories to promote in their
IDPs
Coordinated planning for future water resources interventions for those WSAs supplied by
the WCWSS must be strengthened. Competing applications are evidence that coordination
is lacking. Economies of scale should be considered between WSAs within the Berg to
prioritise the most appropriate (cost effective and resource efficient when considering the
whole) water resources infrastructure, prior to the related WULAs being submitted to DWS
for their consideration. Achieving this does not necessarily require an additional or separate
study to be conducted (by DWS or WCG), it rather requires coordination, conversation, a
platform, and information sharing.
50
5. Value of water to the regional and
municipal economies
5.1. Introduction
The development of a regional hydro-economic model was proposed key intervention 3 of the
project (see Chapter 1). Formal definitions for hydro-economic models are provided in Box 4
below.
Box 4: Formal definitions of hydro-economic models
During the course of the project, it was established, and confirmed through consultation with
various experts and stakeholders, that a hydro-economic model as formally defined could not be
fully realised during this study, primarily due to data constraints.
Instead of a full hydro-economic model, it was proposed that a simpler regional hydro-economic
geographic information systems (GIS) model be developed. This model would have two main
components: (1) the value of water both in economic and social terms, and (2) water
requirements linked to the various economic activities. This information would then be provided
in a spatially explicit manner (GIS) for a particular region - in this case the Berg WMA. Despite
not being as comprehensive or dynamic as the models implied in the formal definitions of hydro-
economic models (Box 4), this approach would enable a consideration of the trade-offs of
different water uses (and hence potential allocations), as both the economic and social value
that the water generates and the amount that is required to generate this value can be
considered in conjunction. A schematic representation of the model components is provided in
Figure 15.
“Hydro-economic models represent spatially distributed water resource systems,
infrastructure, management options and economic values in an integrated manner.”
(Harou et al. 2009)
“Hydro-economic models are based on detailed representation of the river system linked
to relevant economic production activities in the basin through appropriate demand
functions. The demand functions depend on exogenous input–output parameters of the
economic production process and reflect, at best, a partial economic equilibrium system
of demand and supply equations.” (Kimaite, 2011)
51
Figure 15: Regional hydro-economic model components
This chapter presents methodologies for the estimation of the economic and social value of
water to the regional and municipal economies. The estimation of water requirements is
considered in the next chapter (Chapter 6), which also integrates the value of water estimates
from this chapter in order to provide a simplified regional hydro-economic model.
As highlighted in Chapter 4, access to water is a basic human right and water is reserved for the
environmental functioning of a river ecosystem (i.e. the ecological reserve). Water is also a key
input into the production of goods and services that generate economic value. The objective of
water management is to ensure optimum, long-term, environmentally sustainable social and
economic benefit for society (DWA, 2013a). Water allocation is complicated by the fact that
water functions as a connected system. As the system is connected, upstream extractions have
implications downstream. This highlights the need for a spatial, systemic view to ensure that
there is sufficient water to meet the various needs in various areas and provides the motivation
for the geographically explicit approach presented in Chapter 6.
This chapter considers different approaches to estimating the sectoral value of water, utilising
the agriculture sector as a case study (section 5.2). Agriculture is a highly water-intense
industry, and one for which the availability of water is directly linked to economic outcomes
(either through irrigation and/or rainfall). It is also likely to be impacted by climate change and in
highly-contested regions there may be conflict between agricultural and urban water users.
Therefore agriculture provides an interesting case study to illustrate trade-offs between different
water users and the subsequent impact on economic and social outcomes. The estimation of
the value of water for the agricultural sector was thus the focus of one of the Master's studies
funded through this project (Muller, 2017) and formed the foundation for the work on the
sectoral value of water presented here.
5.2. Sectoral value of water: agricultural sector case study
Drawing on economic theory and practice, there would be two approaches to estimate the value
of water for different sectors, namely marginal values and average values. These approaches,
their application and their benefits/drawbacks for the purpose of a regional hydro-economic
model are considered below. Depending on the approach used, different values could be
obtained for the sectoral value of water. For the case of agriculture, previous studies have
estimated the economic value of agricultural water to be R0 - R27.95 per m3 - substantially less
52
than the economic value of water for non-agricultural sectors i.e. R2.40 - R340.57 per m3
(Muller, 2017 p7-8).10
The differences between marginal and average values from a theoretical point of view are
discussed in section 5.2.1, with the estimations of the marginal and average economic value of
water for agriculture detailed thereafter. The benefits and disadvantages of these approaches
for the purpose of estimating the value of water for the simplified regional hydro-economic
model are also presented. An approach to estimating the social value of water on a sectoral
basis is presented in section 5.2.2.
5.2.1. Marginal and average values
Water does not function in an open, free market, but rather its price and its allocation (or quantity
by user) is set institutionally - in part to ensure that the basic human right of access to sufficient
water is met. As the price of water does not reflect the value of water, it cannot be utilised to
inform allocation decisions. Therefore, the value of water needs to be calculated in order to advise
an allocation decision. As indicated, this can be done either through marginal or average values.
The marginal value represents the value of one more additional unit, in this case another unit of
water. The average value, in contrast, is simply the total value divided by the total number of units.
What differentiates the marginal from the average value is that the value changes as the quantity
used changes. The first unit of water allocated usually has a substantial impact, but this value
diminishes as the quantity increases. Whereas, the first unit of water, when valued as an average
value, would provide equal value to the final unit of water used.
Based on neo-classical economic theory, the theoretical ideal approach to inform allocation
decisions would be where the marginal (additional) cost to the user of accessing a unit of water
is equal to marginal (additional) value or benefit gained.11 In an open, free water market, the price
of water would reflect the marginal value and additional water would be purchased until the
marginal value that is derived from the use of water is equal to the price.
Average values, in contrast, would simply consider the value generated per unit of water without
a consideration of whether additional water would increase value. The average value just
provides a general indication of how much value a sector has generated with its allocation. It
may be that there is insufficient demand for a sector’s product and increasing the water
allocation to that sector will result in no increase in value. Using the average value to inform
allocation is thus not ideal as it does not consider the impact an additional value will have, just
what impact such an allocation has had in the past.
10 However, it is important to consider that agriculture receives raw water and not potable water, so, in
practice, the cost of supplying agricultural water is lower than the cost of providing water to households
and industry. 11 Why the marginal cost being equal to marginal benefit is considered the ideal allocation level is best
explained by considering when this is not the case. When the marginal benefit is greater than the marginal
cost, then there is more benefit than cost and the economy could be better off from increasing the
allocation. When the marginal benefit is less than the marginal cost, the economy would have been better
off if this allocation had not been done and then ideally less water would be allocated. Note that this
assumes there are no external costs or benefits from the product i.e. no market imperfections.
53
5.2.1.1. Marginal value of agricultural water
Muller (2017) estimated the marginal value of agricultural water using a complex econometric
model (Tobit regression analysis12). As no field-level water usage data was available, the analysis
was done using representative farms. These were generated utilising municipal level production
and income data from the agricultural census. Census data from different years (1993, 2002 and
2007) was used to create a panel database that could indicate changes in agriculture over time,
which is fundamental to estimating marginal value. This was combined with information on water
use per crop and soil fertility per region (see Muller (2017) for details).
Muller found that the marginal value of water for agricultural products varied significantly by crop,
from R0.14 per m3 for wheat to R4.84 per m3 for peaches (see Table 1 in Appendix 4) for full
details). The marginal value of water also varied significantly by region and soil fertility. This is
clearly illustrated for the case of table grapes for which the marginal value varied from R0.11 per
m3 to R6.66 per m3 by region and R2.96 to R5.76 per m3 according to soil fertility. However, the
data constraints (i.e. absence of field-level data and age of date set13) meant that these marginal
values at best reflect short-term marginal returns (marginal returns change as output changes).
The limitations of this approach is compounded by the fact that many of the agricultural markets
are international, or export facing, and thus their returns are often determined by what is
happening elsewhere in the world. This makes an estimation that only considers a short-term (i.e.
one season’s impacts) of limited use when one wishes to consider the trade-offs between
(allocation of water to) different crops in the longer term. While a theoretically sound approach,
the reality is that decision makers often do not have access to marginal cost and benefit values
and the lengthy process to get these numbers limit their usefulness to decision makers. The
limitations of this approach in the estimation of the value of water based in marginal values was
confirmed when participants at the third workshop stated that these estimates were not useful for
long-term allocation decisions (Appendix 3)
5.2.1.2. Average value of agricultural water
The average value from different crops was considered to inform decisions about the possible
benefit from (both existing and potentially new) water allocations. It was anticipated that these
estimations could be more defensible than marginal values, since more recent data is available
to decision-makers in South Africa, typically from industry associations. In addition, the average
values are more stable over time and thus better suited to longer-term decision making.
In the case of the Berg WMA, there was the additional advantage of being able to generate spatial
explicit information by utilising “fly-over data” i.e. agricultural census data gathered through aerial
photography to identify crop production at a field level and made publicly available in GIS format
(DoA, 2013). Considering water use spatially is important as water resources and water utilisation
are highly spatially dependent (as confirmed, for example, by the variation in the marginal value
of water in different regions in the previous section). As only one year’s detailed field-level data
for agriculture is available from the DoA GIS dataset, it is currently not possible to consider
12 This model considers farmer behaviour in two steps. In the first step, the farmer choses whether to
produce a certain crop on the farm. In the second step, the farmer chooses what share of the farm is used
for the crops that have been chosen. 13 The most recent data points for production and income from an agricultural census is for 2007 i.e. the
data is 10 years old and much is expected to have changed in terms of production and income in the last
10 years.
54
marginal value of water14 using this information, but as additional years of data become available,
this may be a useful avenue for extension of the work in future.
The average value of water was calculated as the value of the crops generated per hectare,
regardless of destination market i.e. as the field level data did not distinguish between intended
market, the value of crops in different markets was ignored and the total value generated per crop
was divided by the total hectares planted. However, the value in different markets can vary
greatly. For example, apricots for export fetch R12,210 per tonne compared to R5,629 per tonne
in local markets (Hortgro, 2012). Using the average value would thus mean that individual fields
would be incorrectly valued. If specific regions have production that is more export (local)
focussed, the approach would under (over) estimate the value that agriculture adds to the region.
The assumption of the average value is also expected to be inaccurate for perennials, that only
produce after a number of years, as the age of the crops is also not considered. This would mean
that fields that are under establishment and not producing any crops are still given the average
production value. This would again have a regional impact if a specific region has a significant
share of newly established perennial crops by over-valuing the value that this region adds to the
economy. However, over time, the perennials will produce and provide economic benefits and
some export orientated producers may be forced to sell their produce locally in certain years.
Over the longer term the value generated would align more with the average values that are
generated, rather than the marginal values which are per definition more short term.
To estimate the average value that irrigated crops generated, a number of databases had to be
consulted as no single database contained information on the value of all crops grown in the
region. The primary data source was industry associations (data for 2011), followed by
government reports, and then international reports and sector-specific reports. For a full list of
data sources see Table 2 in Appendix 4.
The Western Cape flyover data provides field-level data that can be mapped and presented at
different levels of resolution, from the individual field to the entire study area. Utilising this data in
conjunction with the average value of crops, the value that agriculture contributes to different
regions can be calculated. Figure 17 illustrates how the value of irrigated agriculture is distributed
in the Berg WMA. The map highlights the fact that high-value irrigated crops have a concentration
of value in Swartland, Stellenbosch and Drakenstein. This concentration of value is due to a large
number of high-value fields, primarily stone fruit and wine grapes, occurring the region. This is
reflected in the fact that 77% of the irrigated crops in the Berg WMA by area are grapes and 10%
are stone fruit (DoA, 2013). It is important to note that the estimated value is based on average
value per crop which is then mapped onto the fields. The results of the average value per crop by
municipality is detailed in Appendix 4, Tables 3 and 4 and summarised in Figure 16 below. These
values can then be linked to water use as demonstrated in Chapter 6 to form a broader hydro-
economic model that provides the “value of water per drop”.
14 As estimates of marginal value require changes over time to estimate.
55
Figure 16: Average value of irrigated agriculture per year by municipality in the Berg WMA
(R millions)
Figure 17: Value from irrigated agriculture in Berg Water Management Area
56
5.2.2. Social value of water: employment in agriculture
Water is key input into the production of goods and services that generate economic value. Labour
is another key input to production. South Africa struggles with substantial unemployment - the
expanded national unemployment rate was 36,4% in the first quarter of 2017 (StatsSA, 2017).
Employment is thus a fundamental social issue in South Africa that can be indirectly linked to
water use. Employment was thus proposed as an indicator of the social value of water use at a
sectoral level. This section presents a number of alternative approaches to estimating job creation
in the agricultural sector in different regions and for different commodities.
When considering economic value per drop, agriculture performs relatively poorly compared to
other sectors. At a regional level, agriculture contributes only 2% of GVA in the Berg WMA, while
utilising a substantial proportion of the available water in the region. However, this analysis does
not take into account the value that agricultural water use generates in terms of social value. The
agriculture sector’s contribution to employment is especially relevant as it employs a significant
share of low-skilled labour.
To illustrate the importance of employment by the agriculture sector, employment was estimated
using employment multipliers reported by Bureau for Food and Agricultural Policy (BFAB) in
conjunction with the “fly-over data” to develop bottom-up employment estimates (BFAP, 2011 &
DoA, 2013, respectively). This “bottom-up estimate” is compared to the estimates for employment
in agriculture developed by Quantec at the municipal level. This comparison is presented in Table
7 below15.
The bottom-up estimates produce a significantly larger employment number than the regionalised
official statistics. This higher estimate is likely due to the significant seasonal component to
agricultural employment and the undercounting of informal employment (in official statistics).
Informal and seasonal agricultural employment is significant in the agriculture sector and
notoriously difficult to estimate. The only exception, in this case, is Saldanha Bay Municipality,
which has a lower bottom-up estimate. The majority of the crops grown in Saldanha Bay have low
employment multipliers.16 The Saldanha Bay exception also gives weight to the seasonal labour
argument, as the crops in the region are mostly field crops which have a lower seasonal labour
component.
Table 7: Employment in agriculture for municipalities in study area in 2011
Municipality Quantec
estimates** Bottom-up estimate*
Bergrivier 8 104 20 497
City of Cape Town 14 039 13 335
Drakenstein 11 335 35 106
Saldanha Bay 1 582 978
Stellenbosch 6 380 31 948
15 When no crop specific data was available, the minimum value of similar crops was utilised in an attempt
to keep the bottom-up estimates conservative. The assumptions made to make the BFAP multipliers
comprehensive are laid out in Table 3 in Appendix 4. 16 99.7% of the crops by in the Saldanha Bay Municipality, by area, are field crops which have a multiplier
of 0.01 jobs per hectare compared to the multiplier range of 0.75 – 3 jobs per hectare for fruit.
57
Municipality Quantec
estimates** Bottom-up estimate*
Swartland 7 767 27 626
Witzenberg 15 649 38 589
Total municipalities in Berg WMA 64 856 168 079
* Author’s calculations using BFAP (2011) in conjunction with DoA (2013) considering
dryland and irrigated agriculture in the entire municipal boundaries (not exclusively the
Berg WMA).
** Regional standardised employment data from Quantec (2016b)
Agriculture clearly contributes significantly to employment in the rural municipalities, contributing
up to 37% of employment in Bergrivier Municipality in 2011 (Quantec, 2016b). The importance of
differentiating the impact for different regions (municipalities) is emphasised by the fact that
overall agriculture only contributes only 4% of employment in the entire region (Quantec, 2016b).
At a sectoral/commodity level, the bottom-up estimates were compared to commodity level job
estimates by DWS (2017a). This comparison is presented in Table 8. Note that the estimates by
DWS are based only on irrigated agriculture within the Berg WMA, and the bottom up estimates
displayed below are similarly calculated (hence the change in estimates from the Table 7 above)
Table 8: Employment estimates for irrigated crops in the Berg WMA
Crop DWS estimate (2017a) Bottom-up estimate
Grapes - wine 9,039 93,156
Grapes - table 3,736
Pome Fruit 1,283 1,936
Stone Fruit 3,629 8,120
Citrus / sub tropical fruit 989 1,642
Tree Fruit Other 1,140 1,753
Berries 229 855
Grains 88 27
Planted Pasture 103 8
Vegetables 526 1,906
Nuts & oil seeds 1 5
Other - 35
Grand Total 20,762 109,444
The jobs from the fruit sector are significantly larger than those calculated by DWS (2017a). To
examine the validity of these estimates, they are compared to employment statistics recorded by
the industry associations Hortgro (2012) and South African Wine Industry Information and
Systems (SAWIS, 2015) in Table 9. These estimates still seem to indicate a slight over-estimation
in spite of assumptions erring on the lower side (unknown crops and crops without their own
estimates use a minimum of that crop type). This higher estimate may be due to the high amount
of seasonal work that this sector utilises. For example, the South African Table Grape Industry
58
(SATI) (2012) estimates employing 42,505 seasonal and 10,628 permanent employees, i.e.
nearly four seasonal employees for each permanent employee. The seasonal nature of work
could account for this difference.
Table 9: Comparison of fruit sector employment estimates
Industry association estimates Bottom-up estimates
Category Fruit South African Employment
WC share
Western Cape employment
Western Cape employment
% of industry estimate
Pome fruit
Apples 27 493 79% 21 782 34 278
40 613 118%
Pears 14 604 86% 12 496
Stone Fruit
Peaches & nectarines 12 253 88% 10 841
20 804
28 405 137% Plums 6 717 95% 6 352
Apricots 3 819 95% 3 610
Grapes
Dry grapes 39 125 44% 17 301
191 385
196 301 103% Table grapes* 10 628* 62% 6 591
Wine grapes** - - 167 494
Other Citrus - - - - 12 120
Subtropical fruit - - - - 319
Other tree fruit - - - - 3 223
Industry estimates are Hortgro (2012) except for table grapes* which is based on SATI (2012) and only presents permanent employment (excludes 42 505 seasonal workers) and wine grapes** which is based on SAWIS (2015) estimate.
Given the consistently higher employment estimates, the BFAP multipliers were compared to the
jobs estimated through “pseudo multipliers” based on the industry association data. The pseudo
multipliers were calculated by dividing the employment the industry association recorded divided
by the total hectares the associations recorded. The results, shown in Table 10, indicate that the
BFAP multipliers do not seem to be significantly over-stating the employment levels, although
they may contain a component of double counting as labour moves with the seasons. However,
the assumption that wine grapes have a similar labour multiplier to table/dry (raisin) grapes may
be problematic since wine grape production is able to undertake a greater amount of
mechanisation than table grapes.
Table 10: Labour multiplier comparison (jobs/hectare)
Fruit HORTGRO BFAP
Apples 1.25 1.25
Grapes (dry & table) 1.59 1.62
Pears 1.26 1.26
Peaches & nectarines 1.47 1.2 / 1.25
Plums 1.43 1.46
Apricots 1.10 -
59
The labour estimates and multipliers presented here indicate that for agriculture the social value
of water, through the provision of employment, is substantial and may be under-estimated in
official estimates. That said, the amount of uncertainty around the estimates suggests the need
to be conservative, so as not to overstate the social value/job creation from agriculture. However,
the availability of water has been emphasised by agricultural producers as a constraint to
increased production. Therefore, despite the limitations, the average economic value in Rand and
labour multiplier calculations give an indication of what additional value could be unlocked and
which crops would potentially generate more value than others.
Overall, however, both the marginal and average sector-based approaches to the estimation of
the economic value of water and the average sector-based approach to the estimation of the
social value of water presented here are limited, primarily due to a lack of data and hence
uncertainty in estimates. It is thus argued that these approaches cannot at this stage be applied
with a high degree of confidence in the development of a hydro-economic model to support
decision making for water allocation. An alternative approach to the assessment of the economic
and social value of water at a regional and municipal level is thus proposed. The proposed
approach is based on understanding the water intensity of regional and municipal economies,
and the associated risk and hence resilience of these economies to changing water availability.
5.3. Water intensity of regional and municipal economies
5.3.1. Proposed approach to understanding value of water
The calculation of the value of water at a municipal and regional level is facilitated by the fact that
municipalities are typically water service authorities (WSA) and that data on water usage,
economic activity, social conditions etc. are collected at a municipal level. However, while
statistics collected at a municipal level may be done in sub-categories differentiating different
economic sectors, water use categories are not consistently applied by municipalities. This makes
water use comparisons between sectors in different municipalities difficult. This inconsistency is
present in the case of the municipalities in the Berg WMA, but all the municipalities in the area
do, at a minimum, distinguish between residential and other sectors.17
While water use can thus not necessarily be clearly linked to different industrial or commercial
sectors, the value that local economies can attribute to companies that utilise significant amounts
of water can be examined. Examining the economic value and jobs that relate to sectors that have
a high water intensity or medium water intensity without considering the actual water that these
industries utilise in detail gives an indication of local economies’ reliance on water to provide
economic growth and job opportunities. This analysis allows municipalities to consider how
resilient their economies are to possible changes in water availability.
5.3.2. Application to Berg WMA
For the municipalities in the Berg WMA, sectors were categorised according to their water
intensity using the classification of the United Nations World Water Assessment Programme
17 The detailed water use is discussed in detail in section 6.3.
60
(WWAP, 2016) categorising them as low, medium 18 or highly 19 water intense. GVA and
employment data were then sourced from Quantec (2016). This is the same data that is used to
inform the annual Municipal Economic Review and Outlook (MERO) reports produced by the
Provincial Treasure of the Western Cape. MERO reports are intended to be a toolkit to facilitate
decision making by municipalities, government departments, public entities, businesses as well
as national and international organisations interested in investing in the Western Cape. To ensure
that data was comparable, data for the entire municipalities that fall within the Berg MMA were
utilised, even when a share of the municipality fell outside of the study area. This was considered
more appropriate than limiting the consideration to only the Berg WMA portions only, as it is
expected that municipal level decisions would usually be made considering the entire
municipality.
The contribution of sectors to the local municipalities’ GVA and employment was analysed for the
entire Berg WMA in Figure 18 (refer to Tables 7 – 10 in Appendix 4 for details). This shows that
the 10% of economic value (GVA) is contributed by heavily water intense sectors (including
agriculture) with 19% of GVA generated by moderately water intense sectors and the remaining
71% attributable to low water intense sectors. Considering the entire region, the risk of a decrease
in water available for agriculture would seem to have a limited impact on the region’s economy.
However, this is due to the fact that the highly urbanised Cape Town Municipality dominates the
region’s economy, contributing 84% of the GVA of the municipalities that fall within the Berg Water
Management Area.
Figure 18: Gross value added and employment by water intensity of sectors in the
municipalities in the Berg Water Management Area20
18 "Sectors that are moderately water-dependent can be defined as those that do not require access to
significant quantities of water resources to realize most of their activities, but for which water is
nonetheless a necessary component in part(s) of their value chains." WWAP, 2016 19 "Sectors that are heavily water-dependent can be defined as those requiring a significant quantity of
water resources as a major and necessary input to their activities and/or production processes." WWAP,
2016 20 Authors’ calculations utilising Quantec (2016) data and WWAP (2016) definitions. Agriculture is also a
heavily water intense sector. Refer to tables 7 – 10 in Appendix 4.
61
The water intensity of the local economies in the region varies significantly, as shown in Figure
18. The importance of agriculture for certain municipal economies becomes evident when the
municipalities are considered separately. This is most evident for the Bergrivier Municipality,
where agriculture provides 37% of employment. The significance of agriculture as a provider of
employment in the region may be even higher if the bottom-up estimates utilising the relatively
recent DoA (2013) fly-over data in conjunction with employment multipliers from BFAP (2011) are
utilised (i.e. the estimates presented in Table 7).
The difference in water intensity of municipalities outside Cape Town differs significantly between
municipalities. The variation in water intensity could be attributed in part to variation in agriculture,
with GVA for agriculture ranging from 3% in Saldanha Bay to 22% in Bergrivier. The variation in
heavily-water-intense sectors is also notable, varying from as little as 7% in Cape Town (or 9%
outside Cape Town in Drakenstein and Stellenbosch) to as much as 21% in Swartland. The
employment impacts are also important and largely track those of the GVA, with agriculture’s
labour-intensity evident from in its greater contribution to employment. Water availability is
expected to change as a result of climate change (considered in section 6.5). The consideration
of the value of industry linked to different water intensities thus helps to identify where a decrease
in water availability would be of greatest risk to the local economy.
62
Figure 19: Gross value added and employment by water intensity of sectors in the
municipalities in Berg WMA
63
5.3.3. Economic value of water at a municipal level
The economic value of water at the municipal scale was simply split into two categories:
irrigated agriculture and everything else (otherwise referred to as urban). The calculation of
irrigated agriculture has been discussed in the previous sections, and importantly, these values
are directly linked to the fields in which the crops are grown. Therefore, it is possible to clearly
determine how much value was generated within the WMA boundary. The non-agricultural
(urban) value calculations utilised the data from Quantec (2016) on GVA and employment
estimates for the entire municipality (excluding agriculture). For those municipalities that partially
fall outside the WMA, the proportion of the municipality that fell within the WMA was assumed to
be consistent with the proportion of the economy that had generated the economic outcomes
from the water accessed within the WMA. For each municipality, these values were then divided
by the amount of water that the municipal economy used (a detailed description of how these
volumes were calculated is included in chapter 6).
Table 11: Average value of agriculture and urban water in municipalities
Municipality
Value or GVA (R millions) Value R per m3
Irrigated agriculture Urban
Irrigated agriculture
Urban
Bergrivier 653 2,828 16.12 1,114
City of Cape Town 436 350,207 11.02 1,018
Drakenstein 1,909 17,177 13.03 912
Saldanha Bay 14 7,418 31.60 525
Stellenbosch 931 12,778 9.21 934
Swartland 1,030 5,840 8.76 1,004
Witzenberg 355 6,488 15.43 996
Berg WMA 5,327 402,735 11.37 993
The value of urban water seems to be highly consistent with the average value being R993 /m3
except in the case of Saldanha Bay that has a substantially lower value of R525 /m3. This is most
likely due to Saldanha Bay supporting water-intense industry (steel manufacturing and fisheries)
that significantly increases water usage while not increasing GVA to the same extent. Similarly,
the average value of irrigated water is consistent across the region, with Saldanha as an outlier,
likely due to the crops grown there: largely winter grains which require little water but still
contribute economically.
It is clear that that 1m3 of agricultural water adds significantly less to the local economies than the
value from non-agricultural water, but there are a number of caveats to this observation. Firstly,
urban water is mostly potable water, while agricultural water is raw water and there are significant
costs involved in taking raw water to a potable level. This analysis does not take into account the
value that is generated by agricultural produce further down the value chain. Importantly, water is
a binding constraint to more production in agriculture, while for other parts of the economy, this is
not necessarily the case.
64
5.3.4. Social value of water at a municipal level
Access to water is enshrined as a basic human right in the South African Constitution and also
emphasised in United Nations member states’ Sustainable Development Goal (SDG) 6 ‘Water
and sanitation for all’. This is relevant at the municipal level as municipalities are responsible for
the provision of these services to households. It is proposed that, at a municipal level, indicators
linked to achieving SDG 6 be used to assist in informing allocation decisions. This has been
demonstrated in the case of the Berg WMA by Cole (2017). A number of indicators were selected.
The indicators for access to water, access to sanitation and quality of water are shown in Table
12.
Table 12: Proposed indicators for SDG 6.1, 6.2 and 6.3 in Berg WMA
Municipality
1. Water access 2. Sanitation 3. Water quality
Piped water
within 200m
of dwelling
(%)
Piped
water in
dwelling
(%)
Access to
ventilated pit
latrines or
better (%)
Access to
flush
toilets (%)
Drinking
water
quality:
Blue Drop
Score
(%)
Wastewater
quality: Critical
Risk Rating
(CRR / CRRmax
(%)
Bergrivier 99 82 91 90 64 55
City of Cape Town 95 79 93 92 96 49
Drakenstein 99 81 95 94 72 56
Saldanha Bay 99 85 96 96 69 58
Stellenbosch 98 77 91 91 80 80
Swartland 99 79 92 91 74 64
Witzenberg 99 71 93 92 96 39
Berg WMA 97 79 93 92 93 57
South Africa (average) 83 46 70 57 80 72
Source: Cole 2017
The analysis highlights that, although water access is not yet universal, the region is performing
above the South African average on most counts. There were only two areas where
performance was below the South African mean. These are linked to the two water quality
measures, namely Blue Drop score and Critical Risk Rating. The Blue Drop system considers
the performance of water service authorities and their providers via a standardised scorecard. In
the Berg WMA, four municipalities performed below the South African average namely:
Bergrivier, Saldanha Bay, Swartland and Drakenstein. For the Critical Risk Rating on
wastewater treatment, Stellenbosch is the only municipality that has a higher risk that the South
African average. So, in summary, the region performs well in terms of provision of basic
services (water and sanitation), but the water quality in the region can be improved. The fact
that provision is near universal suggests that the priority of water allocation for human
consumption has largely been reached, but still requires some work. It does however allow
some consideration of allocation criteria lower on the National Water Resources Strategy
(NWRS2) (DWA, 2013a) priority list, such as job creation.
65
To assist in determining the social value of water, it is proposed that an estimate of “jobs per
drop” (in real terms “jobs per m3”) be used. To demonstrate this in the case of the Berg WMA,
the Quantec (2016) employment estimates for non-agriculture sectors were considered in
conjunction with the water use estimates (as detailed in Chapter 6) to calculate the job per drop
for non-agriculture sectors by municipality. The agricultural job estimates were generated
utilising the BFAP job multipliers, as previously discussed, and then also by irrigated agricultural
water use (detailed in Chapter 6). These estimates are shown in Table 13. It is clear that
agricultural sector delivers substantially lower jobs per drop than the non-agricultural sectors.
Again, there is a high degree of conformity between the average values, with the exception of
Saldanha Bay, in which lower jobs are generated in agriculture and non-agriculture on average.
This can be explained by the crops grown in Saldanha, largely winter grains, which are highly
mechanised with low employment multipliers. The low job intensity for non-agricultural
production in Saldanha is likely due to the water-intense manufacturing industries that do not
generate proportionally the same number of jobs.
Table 13: Total employment and “job per drop” for agriculture and non-agriculture
sectors in the municipalities of the Berg WMA
Municipality
Employment (number) Jobs per drop
(jobs per million m3)
Agriculture Non-
agriculture Agriculture
Non-agriculture
Bergrivier 6,115 13,825 151 5,446
City of Cape Town 11,340 1,393,581 286 4,050
Drakenstein 33,261 80,833 227 4,290
Saldanha Bay 25 38,007 56 2,692
Stellenbosch 29,265 58,430 290 4,269
Swartland 25,054 28,050 213 4,822
Witzenberg 4,383 34,368 191 5,278
Berg WMA 109,444 1,647,094 234 4,201
It is clear that the non-agriculture sector is the largest employer in all the regions and generally
has an overall lower water consumption in municipalities outside of the City of Cape Town. The
non-agriculture sector thus results in significantly more jobs per drop. However, the non-
agriculture sector includes a wide range of sectors with varying water intensities and it does not
necessarily hold that the job per drop would be higher for all non-agricultural sectors. Given the
inconsistent reporting of urban water usage (in terms of sectoral consumption), a more detailed
consideration of the jobs per drop for different sectors cannot be considered at this stage.
However, this would be a valuable avenue for further research to determine the social value of
water.
5.4. Conclusions
A hydro-economic model was proposed as a means to integrate the allocation of water with its
economic and social outcomes. Although, the marginal value of water is far more preferable
means of estimating this value as it more accurately links the benefit that can be accrued to
changing water allocations, data constraints only allowed for an average value estimation. The
66
analysis of the average value of water in the study area is in alignment with other similar studies,
notably that water utilised for agricultural production produces less value than water used in urban
sectors. This is reflected in both employment and GVA. However, this value is not equally
distributed across the region, with the type of crops grown determining some of this difference. A
structural analysis of the water intensity of an economy was also conducted in order to understand
how resilient it is to water shocks and constraints. This also revealed that local economies in the
region have highly varied water intensities. The combination of the value of water used and the
intensity of this water use is important to consider. A reduction in water supply to a largely
agricultural economy, will not have the regional economic impacts that the same proportional
reduction would have for a largely tertiary economy, like the City of Cape Town, however, the
local impacts would be felt severely. Chapter 6 discusses how water requirements for the region
were calculated and then integrates these requirements with the values discussed in this section,
with the understanding that prioritising water supply to different regions should not only be based
on the total value that this water enables, but also on the relative impact of the water on the local
economy.
67
6. Regional hydro-economic GIS model
6.1. Introduction
In order to better understand how water availability may impact on a local economy’s growth
prospects within the Berg WMA, a regional hydro-economic GIS model21 was developed. The
spatial visualisation of water allocation, overlaid with its economic value (as determined in Chapter
5) provides decision-makers with a clearer view of where water is most likely to be a constraint to
growth, while the comparative value of the water assists in prioritising efforts to close the supply
gap.
The volumetric estimation 22 of water requirements or consumption was broken into two
categories: irrigated agriculture, and urban and industrial use. Dry-land agriculture was not
considered in this study as the analysis focuses on water supply, i.e. water that can be captured,
stored and supplied to end users. Who receives this water supply is institutionally determined and
subject to the National Water Act (amongst other legislation), whereas the water that falls onto
rainfall-fed agriculture is not under the control of the government. Climate change may have an
economic impact on dry-land agriculture (and therefore the local economy), but this would not be
accounted for in this study that focusses on understanding how water demands from different
sectors of the economy may be best realised through different allocation scenarios or supply
schemes.
Water consumed by irrigated agriculture is not currently measured, and the existing database of
Water Authorisation and Registration Management System (WARMS) is acknowledged to be
inaccurate (DWA, 2006) as it is based on a self-registered volume that may not always be
precisely estimated. Furthermore, the practical application of WARMS in geospatial maps has
found the location of many of the farms to be incorrect (i.e. the recorded GPS coordinates indicate
some farms as being located in the ocean). Therefore WARMS was not considered an appropriate
means of estimating irrigated agricultural demand, however, the total regional requirements were
summed and provided a validation check for the bottom-up calculations.
DWS is currently conducting a Validation and Verification process in the Berg WMA (refer to Box
1 in Chapter 4) that will provide an estimate of existing irrigated water consumption. This process
is not yet completed, and will likely be finalised towards the end of 2018. If this data is made
publically available, it will be invaluable in quantifying the approximate consumption of agricultural
users in the region. However, in the absence of this data, a methodology was developed to
estimate irrigated agricultural requirements (section 6.2).
The calculation of urban water requirements was relatively straight-forward as town-level usage
records are available, as described in section 6.3. Combined, this allows forecasts to be
generated for water requirements in 2025 and 2040 utilising assumptions for climate change and
urban growth due to population growth (see section 6.4). The final step in section 6.5 involves
21 Substantial sections of this chapter are derived from the Masters study that was funded as part of this
project (van der Walt, 2017). The methodology employed and data sources used were developed as a
collaborative project team effort. 22 Water requirements/usage are expressed as cubic metres or m3 per year unless otherwise indicated
68
combining the value of water estimates generated in Chapter 5 with the future water requirements
to understand how a lack of water may result in constraints to the economy.
6.2. Estimating irrigated agriculture water requirements
With the increasing availability of spatial and ancillary data, modelling water requirements both
temporally as well as spatially is becoming more and more accessible (Aspinall & Pearson, 2000;
Arnold & Fohrer, 2005; Serra et al, 2016). Water requirement simulations are useful for
investigating the relationship between irrigation management practices and irrigation demand, as
well as to predict future demand with the use of data produced by climate modelling (Arnold &
Fohrer, 2005; Leenhardt et al, 2004; Thomas, 2008).
Crop water requirements can be estimated using a variety of methods. These methods normally
rely on a combination of datasets including data describing local climate conditions over a period
of time, such as daily minimum and maximum temperatures, rainfall volume and intensity, relative
humidity, hours of sunlight and average wind speed, and data pertaining to specific crops and
cultivation methods, such as planting date, length of growth stages, irrigation method, leaf area
index, rooting depth and planting density. Other data often used includes soil type, soil layer water
holding capacity, soil layer depth, surface soil albedo, and susceptibility to saline build-up and
nitrogen leaching (Jensen, 1973; Jensen et al, 1990).
This study models irrigation water requirements simply, using time-series monthly climate data in
conjunction with crop factors. Soil water balance is done to the extent of estimating a moisture
deficit resulting from plant evapotranspiration-reduced soil moisture, replenished only by effective
rainfall and irrigation. No salinity control or other processes are included in the analysis. Focus
was placed only on the water lost through evapotranspiration processes, taking into account
rainwater that entered the soil and was thereby made available to the plant.
A polygon feature class dataset containing field boundaries surveyed by remote sensing (DoA,
2013) was used as a basis for the calculation of irrigation demand. The dataset also provided
important attribute data pertaining to each field’s irrigation regime and crop types. From this data
dry-land fields could be identified and removed as their consumption of green water would not be
considered in this study.
Monthly cumulative effective rainfall over the period between 1979 and 2013 was obtained in the
form of a grid of 0.5 x 0.5 degrees. Downscaled precipitation data was available but was not used
due to limitations in the ability of downscaling methodology to accurately portray daily rainfall
variability (Wilby & Wigley, 1997). Monthly reference evapotranspiration, calculated by the
Penman-Monteith method, was taken from the WRC’s 2007 Agrohydrological Atlas. This data
was used to extract monthly effective rainfall and to calculate monthly reference crop
evapotranspiration rates per field.
In order to investigate the potential range of future irrigation water requirements, mean daily
temperature and effective rainfall data from several General Circulation Models (GCMs) were
investigated. Observed data from 1979 to 2014 was analysed and compared with data from
eleven of the twenty climate models that formed part of the fifth phase of the Climate Model
Intercomparison Project (CMIP5) (Li, 2014; Ji et al, 2014; Chylek et al, 2011; Voldoire et al, 2013;
Bao et al, 2013; Watanabe et al, 2010; Watanabe et al, 2011; Yukimoto et al, 2012; Taylor et al,
69
2012) over the same period to identify models representing drier, wetter and more temperate
trends over the study area.
The following models were evaluated: the Beijing Climate Center Climate System Model (BCC-
CSM1-1), the Beijing Normal University Earth System Model (BNU-ESM), the second
generation Canadian Earth System Model (CanESM2), the Centre National de Recherches
Météorologiques’ GCM (CNRM-CM5), the Flexible Global Ocean-Atmosphere-Land System
model, Spectral Version 2 (FGOALS-s2), the National Oceanic and Atmospheric Administration
Geophysical Fluid Dynamics Laboratory’s Earth Systems Models with General and Modular
Ocean Models (GFDL-ESM2G, GFDL-ESM2M), the Model for Interdisciplinary Research on
Climate and associate Earth System Models (MIROC5, MIROC-ESM, MIROC-ESM-CHEM), as
well as the Japanese Meteorological Research Institute’s Coupled Global Climate Model (MRI-
CGCM3)
6.2.1. Current irrigated water requirements
6.2.1.1. Methodology
In order to map irrigated agriculture in a manner consistent with the land use mapping approach
taken for delineating urban water use zones, a polygon feature class dataset from the 2013
Flyover agriculture survey was obtained, defining the borders of all cultivated fields in the study
area. The dataset was converted into a raster grid coincident with the national land use dataset
used previously, with matching spatial extent and cell sizes, such that the cells within each dataset
would be conterminous.
The resulting agriculture raster grid was then converted to a points feature class producing an
array of points, each located at the centre of a grid cell. Attributes defining the field identification
number, predominant crop type, irrigation method and total field size in hectares were then
assigned to each point based on the properties of the field on which it was overlaid.
Monthly irrigation requirements were calculated at a field level using Penman-Monteith reference
crop evapotranspiration values based on a 50-year climate dataset (Schulze et al, 2007), locally
and internationally derived crop coefficients spanning sixteen different crop categories (see Table
16 below), and effective rainfall calculations.
The Penman-Monteith formula can be summarised as follows (Monteith, 1965):
𝐸𝑇𝑜𝑚 =∆𝑅𝑛+(𝑒𝑎−𝑒𝑑)×
𝜌×𝑐𝑝
𝑟𝑎
𝜆(Δ+𝛾×(1+𝑟𝑠𝑟𝑎))
(1)
Where 𝐸𝑇𝑜𝑚 is the reference evapotranspiration value for a given month, Rn is net radiation
(W/m2), is air density, cp is the specific heat of air, rs is net resistance to diffusion through the
surfaces of leaves and soil (s/m), ra is the net resistance to diffusion through the air from the
surface to the height of the measuring instruments (s/m), is the hygrometric constant, = de/dT,
ea is saturated vapour pressure at air temperature, and ed is the mean vapour pressure. This
method is generally considered to be quite accurate, although rs may become less accurate when
70
a large region is considered, or if vegetation is diverse or distributed in an uneven manner
(Kneale, 1991).
The Penman-Monteith method of estimating reference crop evaporation has become the de
facto international standard. Once this data was extracted from the agro-hydrological atlas
(Schulze et al, 2007) at a field level, crop-specific evapotranspiration levels were determined
using an adapted formula from Allen et al. (1998):
ETm = Kc x ETom
(2)
where ETm is the monthly evapotranspiration value calculated by multiplying the reference
evapotranspiration value for a given month, ETom (mm/day) by the crop coefficient for the
corresponding crop and season, Kc (Table 16) (Mekonnen & Hoekstra, 2011). From this simplified
equation the potential amount of water that could be taken up and released by a plant may be
calculated for each month. Table 14 lists the crop coefficients used in this study.
Table 14: Seasonal time-averaged crop coefficients (Kc) for each irrigated crop type
Crop Types: Reference Crop Coefficient (Kc) Reference:
Summer Dec-Feb:
Autumn Mar-May:
Winter Jun-Aug:
Spring Sept-Nov:
Berries 0.8 0.33 0.2 1 SAPWAT4 (WRC, 2016)
Citrus fruits 0.80 1.28 0.79 0.62 Green and Moreshet (1979)
Grains - - 0.92 0.25 Allen et al (1998)
Grapes 0.95 0.55 - 0.59 SAPWAT (WRC, 2016)
Herbs/Essential oils 0.75 - - 1.15 Allen et al (1998)
Nuts 1.10 0.65 0.50 1.10 Allen et al (1998)
Oil seeds 0.35 - - 1.15 Allen et al (1998)
Other crops 0.35 - - 1.15 Allen et al (1998)
Pepo 0.75 - - 1.05 Allen et al (1998)
Planted pastures 1.25 1.05 0.90 1.06 Allen et al (1998)
Pome fruit 0.77 0.72 0.19 0.46 Taylor & Gush (2014)
Prickly pears 0.47 0.47 0.47 0.47 Allen et al (1998)
Stone fruit 0.85 0.28 - 0.72 SAPWAT (WRC, 2016)
Sub-tropical fruit 0.80 0.70 0.40 0.70 Allen et al (1998)
Tree fruit - other 1 0.5 0.35 0.65 Allen et al (1998)
Vegetables 0.75 - - 1.15 Allen et al (1998)
Crop coefficients were based on three-monthly seasons, and as they were derived from a number
of sources, were adapted to seasons by time-averaging. One of the most important crops in the
Berg WMA is grapes. The crop coefficients for grapes were captured from SAPWAT4 (WRC,
2016) with the following parameters:
Plant date: 1 September
Bud Break: Early Spring
Climate: Dry/Cold
71
Table 15: SAPWAT4 Wine Grape growing season and corresponding crop coefficients
Initial Developing Mid Late Total days
Days per period 40 80 105 45 270
Kcb by growth period 0.15 0.95 0.95 0.15
Kc = Ke + Kcb where Kc is the crop coefficient applied, Ke is the soil evaporation coefficient and Kcb is basal crop coefficient. The Ke value was found to have little impact on the overall annual irrigation estimates, as the soil type only appeared to have effect in the start of the season (see Figure 19 below), so when crop coefficients from SAPWAT were utilised, Kcb was considered analogous to Kc.
Figure 19: Screenshot from SAPWAT4 (WRC, 2016) of wine grape irrigation estimate
The days of each growing period were overlaid with the corresponding season, where the days
did not exactly correspond to the season boundaries, the coefficients were proportionally
calculated:
Table 16: SAPWAT4 Wine grape crop coefficients by season
Season Spring Summer Autumn Winter
Month Sept Oct Nov Dec Jan Feb Mar Apr May June July Aug
Monthly Kcb 0.15 0.68 0.95 0.95 0.95 0.95 0.95 0.55 0.15
Seasonal Kcb 0.59 0.95 0.55 -
Effective rainfall (ER) was calculated using the U.S. Bureau of Reclamation Method (Stamm,
1967) (Table 17) and treated as a displacing factor in irrigation demand.
72
Table 17: Effective rainfall estimates based on accumulated rainfall using the U.S.
Department of Reclamation Method (Adapted from Adnan & Khan, 2009)
Precipitation Increment Range (mm)
Percentage of Rainfall Considered Effective
Effective Precipitation Accumulated Range (mm)
0.0 – 25.4 90 – 100 22.9 – 25.4
25.4 – 50.8 85 – 95 44.4 – 49.5
50.8 – 76.2 75 – 90 63.5 – 72.4
76.2 – 101.6 50 – 80 76.2 – 92.7
101.6 – 127.0 30 – 60 83.8 – 107.9
127.0 – 152.4 10 – 40 86.4 – 118.1
> 152.4 0 – 10 86.4 – 120.6
The difference between the monthly effective rainfall (ERm) and the monthly evapotranspiration
(ETm) represents the moisture deficit to be covered by irrigation, which is then converted into
monthly irrigation requirement:
𝐼𝑚 = 100(10A((𝐸𝑇𝑚−𝐸𝑅𝑚)
𝐸𝐹), 𝐼𝑚 > 0
(3)
where Im is the monthly irrigation requirement, A is the field are in hectares, and EF the irrigation
efficiency (Table 18). Negative values were treated as zero irrigation requirements as it was
assumed that plants cannot utilise water beyond the point at which potential evapotranspiration
is reached.
Table 18: Irrigation types found in the study area with relative efficiencies (DoA, 2013 &
Ascough & Kiker, 2002)
Irrigation Type Efficiency
% Number of
fields Hectares
Dragline irrigation 75 1 9
Drip irrigation 85 27,240 69,804
Flood irrigation 65 1 16
Floppy irrigation 85 1 2
Pivot irrigation 85 210 4,084
Sprinkler irrigation 75 339 994
Total 27,792 74,910
6.2.1.2. GIS Model
Once the monthly irrigation requirement for each field in the 2013 agriculture census flyover
dataset had been calculated in Excel, the field polygons were converted to grid-based cells,
coincident on the land use grid used for the spatial delineation of urban water use. A per-cell
irrigation requirement was then calculated per field. Once the per-cell values were calculated for
irrigated land uses, each point was assigned a water requirements value based on its land use
type and location (see the PAST23 attribute field in Table 19).
23 PAST simply refers to estimates of irrigation requirements on the basis of historical climatic conditions
73
Table 19: Irrigation requirements database sample
FIELD_ID 18260 42201 112516
FARM_ID 576 38735 49597
MUNIC Swartland Stellenbosch Drakenstein
DISTRICT West Coast Cape Winelands Cape Winelands
CATNAME Viti Viti Viti
AREAHA 3.52 1.14 0.65
CAPDATE 22-10-13 May / June / July 2013 May / June / July 2013
CROPS Wine grapes Wine grapes Table grapes
CR_DETAIL Wine grapes (3.52) Wine grapes (1.14) Table grapes (0.65)
CR_SUM Grapes Grapes Grapes
DRY_IRR Irrigated Irrigated Irrigated
IRR_TYPE Drip irrigation Drip irrigation Drip irrigation
PAST 23684.0 6538.0 4261.0
The irrigation water requirements database outputs are linked to the Western Cape Flyover
polygon shapefile through a process called “Join” within ESRI ArcGIS which joins a layer to
another layer or table based on a common field. The records in the Join Table are matched to the
records in the input Layer Name. A match is made when the input join field and output join field
values are equal. The common field used for this matching was the Field ID, present in both
databases.
Figure 20: Overview of irrigated water requirements GIS model
After the Join has been completed one is left with an irrigation water requirements database (1)
in the form of a shapefile. Particular municipalities or the entire WMA can be selected for visual
analysis in Step 2, as seen above.
A shapefile is created at a field level and added to the display. Data is then dissolved24 in order
to conduct a spatial join to the municipal shapefile. As part of the dissolve process (3), the
aggregated features can also include summaries of any of the attributes present in the input
features (refer to Table 19 above). For instance, the total water requirements of a particular
24 In GIS, dissolve is one of the data management tools used for generalizing features. It is a process in
which a new map feature is created by merging adjacent polygons, lines, or regions that have a common
value for a specified attribute.
74
municipality selected in step 1 is summed to give the total irrigation water requirement for the
selected municipality.
The polygon shapefile is then converted to a point feature class (4, 5) to assist in the spatial join
operation in step 6 of the process. This is necessary as the polygons can overlap into more than
one municipality at a time.
The final step in the process is a spatial join (6) to the surveyor general municipal shapefile (5).
A spatial join involves matching rows from the Join Features to the Target Features based on
their relative spatial locations. By default, all attributes of the join features are appended to
attributes of the target features and copied over to the output feature class, which is the municipal
layer.
The outputs of this process produces shapefiles that can be utilised in a GIS environment for
visual analysis as well as further database queries.
Figure 21: Screenshot of shapefile outputs for irrigated fields
6.2.1.3. Results
The results from the bottom-up estimations of irrigated water requirements are summarised in
Appendix 5 and in Figure 22 below:
75
Figure 22: Irrigated agriculture water requirements per year in the Berg WMA
This graph (Figure 22) illustrates how grapes, largely wine grapes, consume most of the
irrigated water in the region – estimated at approximately 79%. Drakenstein, Swartland and
Stellenbosch are the largest irrigated agricultural water consumers and these are predominantly
wine growing regions. The map below (Figure 23) also illustrates the variability of irrigated water
requirements by municipality across the WMA
76
Figure 23: Irrigated water requirements per municipality per year
In terms of water intensity, grapes require an average amount of water per hectare in
comparison to the other crops grown in the region, as seen below (Figure 24).
77
Figure 24: Average water requirement per hectare per year
However, the water intensity of grapes varies across the region, due to the local climatic
conditions, as shown in the map below (Figure 25). With increased water scarcity, crop
switching could occur, as farmers look to plant more water efficient crops in those regions where
certain crops may produce marginal value in comparison to the water they consume. However,
this assumption is largely dependent on farmers having the information readily available to
understand the trade-offs between water intense crops and the revenue that these commodities
could earn.
78
Figure 25: Grapes water intensity per hectare per municipality
6.2.1.4. Validation of bottom-up estimates
The validation of the irrigated agriculture water requirement estimates generated in this project
was done on two levels with a number of different sources:
At a macro scale where annual estimates for a region were compared with other models
On a field by field basis where samples were taken and compared
A margin of error in estimating the bottom-up calculations was expected. Crops were grouped
into categories that don’t take into account the variability of different cultivars on water
consumption. Factors such as soil type were also not considered, but could have an impact at the
start of a rainy season due to variability in the run-off. Furthermore, the expected irrigation
requirements do not mirror actual farmer behaviour, for whom heuristics and previous experience
may drive irrigation performance.
Crop coefficients are noted as the most significant contributing factor to inaccurate irrigation
estimates (WRC, 2016). Therefore the crop coefficients utilised in this study should be carefully
scrutinised and updated according to new information on region or cultivar-specific crop
coefficients. DWS uses SAPWAT for generating estimates of irrigation requirements as part of
their Validation and Verification process (DWAF, 2006) and SAPWAT was utilised to generate
crop coefficients where it was discovered that the crop water requirements did not meet the
validated results.
79
For a detailed description of the validation steps undertaken, refer to Appendix 6, with summary
results displayed below:
Validation at a regional level:
Water Authorisation Management System (WARMS) records for the Western Cape were
spatially located and the agricultural registrations that fell within the Berg WMA were totalled
Aurecon’s Water Resource Simulation Model (WRSM) of the Upper Berg irrigation demands
were compared to a clipped study area.
Table 20: Comparison of project estimations of irrigated water requirements at regional
level to other available models
Berg WMA Project
Estimations
Berg WMA WARMS
records
Upper Berg River
Clipped Project
Estimations
Aurecon WRSM
estimations for the
Upper Berg River
Total Irrigation
Requirements m3/a 468,556,760 438,932,728 201,135 325 239,000,000
For the Berg WMA, the project estimates are slightly higher than those recorded through WARMS.
This is not unexpected due to the possibility of farmers under-registering their water usage, as
well as the noted issues with the accuracy of the WARMS records. The Aurecon model estimates
are slightly higher than the project estimates for the Upper Berg region. This may be due to factors
utilised in the WRSM, such as evaporation from irrigation canals, that are not included in the
project model. However, these comparisons indicate that the project estimates are largely in
alignment with other regional models.
Field by field validation
Sampled field irrigation estimates were calculated in SAPWAT4, matching the field location
in the same quaternary catchments, with sensitivity analysis conducted for soil type
Sampled project fields were matched to fields within Fruitlook, an online platform that
captures satellite-based information on crop growth and water usage, to assist fruit and wine
farmers with their irrigation scheduling
The field by field comparisons indicated which crop coefficients were providing inaccurate
regional estimates, and were updated accordingly.
6.2.2. Future water requirements
6.2.2.1. Methodology
The estimation of future water requirements for irrigated agriculture were generated for the years
2025 and 2040. The basis of the existing water requirements for agriculture were based on crop
type, irrigation type and climatic conditions. The crop and irrigation type were assumed to remain
constant, but the climatic conditions updated for 2025 and 2040 on the basis of General
Circulation Models (GCMs) sourced from UCT’s Climate System and Analysis Group (CSAG).
The climate models included mean monthly effective rainfall and temperature data from
01/01/1960 to 31/12/2099.
80
The following models were evaluated: the Beijing Climate Center Climate System Model (BCC-
CSM1-1), the Beijing Normal University Earth System Model (BNU-ESM), the second generation
Canadian Earth System Model (CanESM2), the Centre National de Recherches Météorologiques’
GCM (CNRM-CM5), the Flexible Global Ocean-Atmosphere-Land System model, Spectral
Version 2 (FGOALS-s2), the National Oceanic and Atmospheric Administration Geophysical Fluid
Dynamics Laboratory’s Earth Systems Models with General and Modular Ocean Models (GFDL-
ESM2G, GFDL-ESM2M), the Model for Interdisciplinary Research on Climate and associate
Earth System Models (MIROC5, MIROC-ESM, MIROC-ESM-CHEM), as well as the Japanese
Meteorological Research Institute’s Coupled Global Climate Model (MRI-CGCM3).
Without any degree of certainty as to which climate model best represented the future climatic
conditions in the region, three models were selected from the possible 11. These models were
selected on the basis of the sensitivity analysis outputs they provided:
FGOALS: high temperature, low rainfall (i.e. worst case scenario)
GFDL-ESM2G: low temperature, high rainfall (i.e. best case scenario)
CANESM2: closest correlation for past temperature observations (i.e. closely correlated
scenario to existing trends)
For each GCM, low and high emission scenarios were also considered. The 2025 and 2040 were
averaged on a 10-year mean. Below are the temperature and effective rainfall comparisons
between the climatic models used, including observed historical data (1979 – 2014).
Figure 26: Mean monthly temperatures in the Berg WMA
81
Figure 27: Mean monthly effective rainfall in the Berg WMA
As can be seen, all the climate models predict higher temperatures than the currently observed
trends and this will increase the evapotranspiration rates of the crops. However, the rainfall
predictions are much difficult to consistently analyse, but the trends do show a shift in rainfall
patterns with later rain expected in autumn and spring according to current trends. Yet, it is not
clear if the annual rainfall volumes will be similar to current volumes.
The GCMs had been down-scaled and were modelled on a quadrant basis at the scale of
45kmx55km.
82
Figure 28: GCM climate quadrants for the Berg WMA
For each quadrant in the figure above, temperature and rainfall forecasts had been estimated and
from that, the crop water requirements were calculated. Due to the number of measurements
involved in calculating the Penman-Monteith reference crop potential evapotranspiration rate
(PE), an analogous formula was used to estimate future crop water requirements, based on
average temperatures and effective rainfall values. The Thornthwaite method (Thornthwaite,
1948) (Equation 4) was chosen due to its relative simplicity:
𝑃𝐸𝑚 = 16𝑁𝑚 (10�̅�𝑚
𝐼)𝑎
(4)
where m denotes the respective month, Nm is the adjustment factor representing hours of daylight,
Tm is the monthly mean temperature, I is the heat index for the year:
𝐼 = ∑ 𝑖𝑚 = ∑(𝑇𝑚̅̅ ̅̅
5)1.5
(5)
and: 𝑎 = 6.7 × 10−7 × 𝐼3 − 7.7 × 10−5 × 𝐼2 + 1.8 × 10−2 × 𝐼 + 0.49 (6)
A baseline comparison dataset was calculated to determine how the simpler Thornthwaite formula
would compare with the Penman-Monteith method for calculating evapotranspiration. The
baseline reference crop potential evapotranspiration rates were used to calculate a simple
calibration factor in order to adjust PE values based on projected future climate conditions:
83
𝐹𝑐𝑚 =𝐸𝑇𝑚
𝑃𝐸𝑚
(7)
Figure 29 shows the distribution of the resulting calibration factors. New irrigation requirements
were then calculated based on the adjusted PE values, using the crop factors and methodology
described above.
Figure 29: Distribution of average monthly ETm /PEm ratios
The monthly average ratios between the Penman-Monteith and Thornthwaite
evapotranspiration figures show a distinctly seasonal trend, with an outlier in the April average
ratio. September represents the month where the ETm value is largest relative to the PEm value,
while March represents the month where the two estimates are closest. The calibration factors
were not as expected: the significant rise in April, as well the quantum of the difference (around
150% higher on average). Numerous checks were conducted on the data and calculations, and
no errors observed, however this should be investigated further if there is a concern. Important
to note is the scale at which the two models are estimated: the Agrohydrological atlas provides
climatic data at 30mx30m, while the down-scaled GCMs are at 45kmx55km.
Figure 30 shows a comparison between the reference crop ET figures obtained from the South
African Agrohydrological Atlas, calculated using the Penman-Monteith method, and PET values
calculated using the Thornthwaite method. The ET values were calculated from temperature
averages based on a 50-year dataset from 1950 to 1999, while the PET values were calculated
based on temperature averages over a 35-year period from 1979 to 2013.
1.491.41 1.36
2.25
1.41
1.55
1.70
1.861.94
1.821.76
1.56
0.00
0.50
1.00
1.50
2.00
2.50
J A N F E B MA R A P R MA Y J U N J U L A U G S E P O C T N O V D E C
84
Figure 30: Comparison of historical monthly averages for potential reference crop
evapotranspiration over the study area calculated using the Penman-Monteith method
over a 50-year period (1950 – 1999) and the Thornthwaite method over a 35-year period
(1979 – 2013).
While the different temporal spans of the two datasets were anticipated to result in different
monthly averages, the different trends suggest an explanation for the distribution of the
calibration factors for the month of April (Figure 29) but not the overall calibration quantum.
6.2.2.2. Results
The results for future irrigation requirements are summarised in Appendix 7, according to the
total irrigated water requirements by municipality, by climatic model utilised. The results for the
entire Berg WMA are summarised below according to the climatic model utilised, the emissions
scenario assumed (either low or high) and the year 2025 or 2040. The x-axis on the graph
below should be read as follows: the GCMs are named according to their organisation, i.e.
GFDL, FGOALS and CANEMS2. The first number, 45 or 85, reflects the emissions scenario. 45
is the low emission scenario, and 85 the high emission scenario. The second number is the
year with 25 reflecting the estimations for 2025 and 40 for 2040.
85
Figure 31: % increase in irrigated water requirements for Berg WMA in 2025 and 2040
The results reflect that regardless of the model utilised, increased irrigated water requirements
are expected across the region. There is also consistently an increase between 2025 and 2040
expected, and the higher emissions scenarios result in higher requirements than the lower
emissions scenarios. Interestingly CANEMS2 (the best-correlated model with past results)
estimated the lowest increase in requirements in 2025, with very similar results to GFDL in
2040. FGOALS (worst case scenario), as expected, estimated the highest irrigated
requirements for all years and emission scenarios. It was decided to utilise CANEMS2 45 as the
“business as usual” climate model for use in estimating total water requirements for the region
as this model was best correlated with existing records and produced the most conservative
impacts on irrigated agriculture.
The results by municipality vary significantly, due to the local climate forecasts and the crops
grown in that municipality. Using the results from CANEMS2 45, it appears that Saldanha Bay
should expect the biggest increase in water requirements, with Witzenberg the lowest increase.
However, important to note that Saldanha’s requirements under different GCMs differ
significantly, with decreased water requirements expected for GFDL and FGOALS. This is
largely due to grains being the dominant crop grown in that region. Grains have a short growing
season and are impacted by the timing of rainfall, which fluctuates between models (see Figure
27 above).
86
Figure 32: % increase in irrigated water requirements by municipality in 2025 and 2040
using CANEMS2 45 climate model
However percentage increases hide the total impact of changing climate on the total water
irrigated requirements for a municipality. Utilising CANEMS2 45, the maps below illustrate how
Drakenstein, Stellenbosch and Swartland continue to require the largest volume of irrigated
water in the Berg WMA in 2025 and 2040.
87
Figure 33: 2025 estimated irrigated water requirements by municipality
Figure 34: 2040 estimated irrigated water requirements by municipality
88
The forecasted changes in irrigated water requirements also differ by crop, this is due to the
relative impact of changing crop evapotranspiration estimates as a result of increased
temperatures, as well as changing rainfall patterns and their link to the growing season of the
crop. The increases below are reflective of the total region’s requirements and therefore the
change expected by crop is also linked to where that crop is predominantly grown and their
climatic changes.
Figure 35: % increase in irrigated water requirements by crop in 2025 and 2040 for the
Berg WMA
Grains are an outlier in this climate model (notably this is also the only crop that decreases its
water requirements between 2025 and 2040), and represent a small proportion of the water
requirements in the region (0.2%), and are therefore removed from the below analysis
89
Figure 36: % increase in irrigated water requirements by crop in 2025 and 2040 for the
Berg WMA (excluding grains)
These results highlight that, unfortunately, the most important crop in the region, grapes, is
expected to increase their requirements substantially by 2040 (34% from current estimates).
Stone fruit, the second most prevalent crop, is also expected to increase its irrigated
requirements by 2040, although to a lesser degree (29%). Citrus fruits, which is the third largest
crop in the region, and it also relatively water intense, is also expected to increase its
requirements significantly (35%). This highlights how crop switching to more water efficient
crops should be expected in this region, if no additional water allocation to agriculture is
secured. At the same time, farmers should be encouraged to invest in water efficiency
technology to ensure the longer-term viability of their farms.
6.3. Assessment of urban water requirements
The assessment of non-agricultural water requirements, termed urban water requirements25, is
particularly important for this WMA. Nationally, agriculture uses approximately 67% of the
available water resources, whereas for the users who are connected26 to the WCWSS, the
proportion of water use by urban communities is much higher – approximately 65%. The urban
population in this region is also growing rapidly, driven by better job prospects in urban centres
as well as the relatively strong performance of the Cape Town and Western Cape economies in
25 Urban water requirements refer to all users supplied by a municipal system, including farms that are
connected to the municipal network for potable water use. This category also includes industrial users
outside of urban areas that are abstracting raw water. 26 The irrigated agricultural water requirements estimated in the section above included farms connected to
the WCWSS as well as other farms that are irrigating using other sources of water.
90
comparison to most other regions in the country (excluding Gauteng). Urban areas are drivers
of economic growth in the region, therefore securing water supply for these areas is going to be
crucial for their future development prospects.
Figure 37: WCWSS allocations by user type (GreenCape, 2017)
6.3.1. Current urban water requirements
6.3.1.1. Methodology
The calculation of urban water requirements is comparatively simpler than irrigated agricultural
requirements, as this usage is measured and reported on by municipalities to DWS. For
industrial users outside of municipal areas, where records of actual usage were not available,
WARMS registration data was utilised. 2011 was the year utilised as the baseline, effectively
two years prior to the fly-over data that formed the basis for the irrigated agricultural
requirements. 2011 was the last national census, and records from the municipalities on water
usage were the most comprehensive in 2011/12.
Town data
Worley Parsons assists municipalities in the region with the annual water services audit reports
that are delivered to DWS as part of their compliance requirements. This is public data, but it is
not easily available, however it was secured through the assistance of Worley Parsons. Each
town within the WMA was geo-located and included if it fell within the WMA boundaries. For
each town, the consumption records for 2011 were recorded. These records included
categories of water use, for examples, residential, commercial and municipal. However the
application of these categories was not consistent between towns and therefore discarded.
Total system input volume was recorded, as well as the population figures for the town from the
2011 Census. From this, per capita consumption for each town in 2011 was calculated by
dividing total annual water usage by the population in the town.
91
Table 21: 2011 Town water usage and population figures
Municipality Town
2011 usage
m3/a
2011
population
2011 per capita
usage m3/a
Bergrivier Porterville, De Lust, 495,947 7,107 69.8
Bergrivier Velddrif 1,005,660 11,027 91.2
Bergrivier Dwarskersbos 83,274 668 124.7
Bergrivier Aurora 41,072 576 71.3
Bergrivier Piketberg, Wittewater 821,477 12,075 68.0
Cape Town Cape Town 326,271,703 3,684,468 88.6
Drakenstein Saron 594,291 7,814 76.1
Drakenstein Gouda 159,918 2,985 53.6
Drakenstein
Paarl, Wellington, Simondium,
Water-Vliet, Val De Vie 16,820,331 197,567 85.1
Saldanha Bay Hopefield 358,670 6,212 57.7
Saldanha Bay StHelena 1,698,930 11,457 148.3
Saldanha Bay Langebaan 1,368,254 8,295 164.9
Saldanha Bay
Vredenburg, Jacobsbaai,
Paternoster, Louwville 4,948,844 40,805 121.3
Saldanha Bay Saldanha 5,659,283 27,915 202.7
Stellenbosch
Stellenbosch, Elsenburg,
Raithby, Lynedoch 10,273,397 82,966 123.8
Stellenbosch
Pniel, Kylemore, Groot-
Drakenstein, Dwarsrivier 666,729 6,893 96.7
Stellenbosch Franschhoek, La Motte, Groendal 1,472,275 18,707 78.7
Stellenbosch Klapmuts 376,656 7,574 49.7
Swartland Riebeek West 214,216 4,633 46.2
Swartland Riebeek Kasteel 263,467 5,431 48.5
Swartland Yzerfontein 320,386 1,132 283.0
Swartland Darling 699,043 10,477 66.7
Swartland Moorreesburg, Klipfontein 762,024 12,818 59.4
Swartland Koringberg 59,952 1,214 49.4
Swartland
Malmesbury, Chatsworth,
Abbotsdale, Kalbaskraal 3,146,105 45,502 69.1
Witzenberg Tulbagh 833,100 9,457 88.1
Rural data
Outside of the towns in the region there are a few industrial water abstraction points. These
were recorded using the WARMS dataset, which includes each abstraction point’s coordinates,
the water source, the registered annual abstraction volume, as well as administrative details
pertaining to the registration and nature of the abstraction record. For municipalities that only
partly fall into the Berg WMA boundaries, the rural population was estimated using the following
method: Total municipal population less the Towns population (both within and outside the Berg
WMA) multiplied by the proportion of the municipal area that falls within the Berg WMA. In a
similar manner to the Towns calculations above, the per capita consumption was calculated
using the industrial water usage divided by the rural population. For some municipalities it was
92
assumed there was no rural population: City of Cape Town – all residents within the metro
boundaries and therefore likely to be included in Census for urban populations. However,
WARMS industrial abstraction points were included in the urban calculations. Witzenberg only
has the town of Tulbagh supplied by the WCWSS and although Wolseley falls on the boundary
of the Berg WMA, it was not included. Witzenberg has such a small proportion of its total
municipal area within the WMA, and only a single WARMS industrial point was recorded near
the town, so Witzenberg only consists of the town of Tulbagh.
6.3.1.2. Results
Refer to the full results in Appendix 8. The City of Cape is responsible for 85% of the total urban
water usage in the Berg WMA, which is very close to the contribution that the City makes to the
total region’s GVA (84%). The dominance of the City in this region implies that any regional
change in water usage from urban populations is going to be driven by the City.
Figure 38: Urban water usage by municipalities in the Berg WMA
This importance of the City of Cape Town’s water usage in the region is well illustrated in the
below map (noting the step change in the legend to accommodate just how much more the City
uses in comparison to the other municipalities).
93
Figure 39: Total urban water requirements per municipality in 2011
The per capita consumption by town is displayed in the graph below. Yzerfontein in Swartland is
a clear outlier, but not unusual for a holiday town that experiences high holiday-goer influx
during peak season, but that has a small permanent population. Saldanha Bay Municipality has
a number of high consumption towns, this is due to the industrial water consumption in the town
from steel manufacturing and the fisheries. The average per capita consumption for the region
is approximately 90 m3/capita/a or 247 litres/capita/day – this is 16% higher than the national
average of 208 litres/capita/day (Cole et al 2017).
94
Figure 40: 2011 Per capita urban consumption per town (m3/capita/a)
6.3.2. Future urban requirements
6.3.2.1. Methodology
The estimation of future urban water requirements in 2025 and 2040 was simply based on an
extrapolation of population growth rates, with an assumption that per capita usage would remain
constant. For each town listed in the Berg WMA, historical population growth rates (2001 –
2011) were extracted from the town’s second reconciliation strategy (DWS, 2015). In addition,
the low, medium and high growth rates were also captured to provide for sensitivity analysis,
should growth rates change in these towns. For the rural populations, the overall municipal
growth rates were applied.
The historical growth rates (2001 – 2011) were based as the “business as usual” baseline to
estimate future total requirements. However, in mid-2017, the 2016 Community Survey results
were released by Statistics SA (Stats SA, 2017). This provided a significant update to the
previous population growth rates seen between the last two censuses. The Community Survey
data was not available at the town level, but it had been split into urban (All settlements) and
rural (Non-Urban) by municipality. In order to provide the most up to date population figures for
2025 and 2040, the 2016 data was then used, which required some manipulation of the
previous urban water requirements calculations. The town water usage for 2011 per
municipality was totalled and divided by the town population to generate an urban per capita
usage. The rural estimates were generated in a similar manner.
The difference between the 2011 and 2016 population growth rates is quite marked, with every
municipality seeing a decrease in population growth in 2016 (see Table 22 below). This is
95
important to incorporate into the modelling of future requirements, as population growth rates
compounded over a long period of time can grow quite dramatically, and impact on the results
(see the following section) of future urban water requirement estimates significantly. Some of
the municipal estimates may be not as accurate, as the 2016 data is not available at a town
level, but this was counter-balanced with the larger concern for more recent population growth
figures. The City of Cape Town’s -46% change in the population growth rates between 2011
and 2016 was also a large driver to consider these updated figures, with 85% of the regional
population.
Table 22: Comparison between 2011 and 2016 population growth rates by municipality
(Stats SA, 2016)
Municipality 2001 - 2011 %
growth rates
2011 – 2016 %
growth rates
% change
between 2016
and 2011 growth
rates
Bergrivier Urban 3.10 1.90 -39%
Rural 2.50 1.50 -40%
Cape Town Urban 2.60 1.40 -46%
Drakenstein Urban 3.20 2.60 -19%
Rural -0.30 -0.30 0%
Saldanha Bay Urban 3.70 2.40 -35%
Rural -0.40 -0.30 -25%
Stellenbosch Urban 3.20 2.40 -25%
Rural 1.50 1.30 -13%
Swartland Urban 4.70 3.30 -30%
Rural 4.60 3.30 -28%
Witzenberg Urban 2.80 -0.80 -129%
6.3.2.2. Results
The results by municipality for the estimated urban water requirements in 2025 and 2040 are
included in Appendix 9 and summarised below:
96
Figure 41: % change in urban water requirements in 2025 and 2040 per municipality
according to historical population growth rates
The water requirements for the City of Cape Town grow at the slowest rate (excluding
Witzenberg), yet still accounts for 80% of the regional usage by 2040, down from 85% in 2011.
The updated 2016 population growth figures highlight Swartland as a rapidly growing
municipality, closely followed by Drakenstein and Saldanha Bay. In the absence of significant
new supply options, reducing per capita consumption through Water Conservation and Demand
measures is going to become increasingly important for these municipalities. The City of Cape
Town has managed to de-couple its population growth from their water consumption trends for
the past decade or so (see below).
Figure 42: Water treated per year and population in the City of Cape Town
97
The summary growth rates for the region (see Figure 43 below) provide further impetus for
using the updated 2016 population growth figures, as these produce more conservative
estimates than even the low growth scenarios outlined in the All Town Reconciliation Strategies
for the towns in the region.
Figure 43: Urban water requirement growth rates for the Berg WMA under different
population growth scenarios
The updated maps of the region for 2025 and 2040 below, continue to highlight how Cape Town
is by far the largest user of urban water in the region, with the other municipalities starting to
emerge as higher users by 2040.
98
Figure 44: Urban water requirements per municipality in 2025
Figure 45: Urban water requirements per municipality in 2040
99
6.4. Total water requirements
6.4.1. Current total water requirements
The total water requirements were calculated by adding the urban (or non-agricultural
requirements) to the irrigated water requirements (Table 23). Noting that these requirements
were calculated at slightly different times (2011 and 2013), but with the assumption that the
crops grown in the region in 2013 would be similar to the 2011.
Table 23: Total annual water requirements for the Berg WMA
Municipality 2011 urban water
requirements m3/a
2013 irrigated
agriculture water
requirements m3/a
Total current water
requirements m3/a
Bergrivier 4,598,070 40,489,366 45,087,436
Cape Town 332,352,881 39,587,871 371,940,752
Drakenstein 18,933,029 146,462,393 165,395,422
Saldanha Bay 14,077,231 451,297 14,528,528
Stellenbosch 13,560,881 101,060,066 114,620,947
Swartland 7,686,201 117,502,212 125,188,413
Witzenberg 877,746 23,003,556 23,881,302
Total 392,086,039 468,556,760 860,642,799
The pie chart below (Figure 46) is a significant update to the breakdown between urban and
agricultural water usage seen on the WCWSS, with urban usage falling below the majority
(WCWSS urban allocations are 65%). This illustrates that once all water usage is taken into
account, not just for those farms that are supplied by the WCWSS, how agriculture is still a large
water user in this region. And a reminder, that this study has only included irrigated agriculture,
dry-land agriculture has not been calculated.
Figure 46: Total water requirements for the Berg WMA in m3/a
100
However, as noted in chapter 5, the water intensity of a local economy varies by municipality,
and this is reflected in the split between agricultural and urban usage in the graph below. The
City of Cape Town continues to consume the most amount of water, but the gap is not as
dramatic as previously, with the more rural municipalities requiring substantial volumes of water
for irrigation (refer to Figure 47 below).
Figure 47: Total water requirements by municipality in m3/a
101
Figure 48: Total annual water requirements by municipality
6.4.2. Future total water requirements
In a similar manner to the current total water requirement estimates, the urban and agricultural
totals for 2025 and 2040 were summed in order to estimate future water demands from the
region (refer to Appendix 10 for the results). This reveals that the driver of increased water
demands in the region is going to be from urban growth, not agriculture, despite the concerning
climatic predictions for the region and the high proportion of water being used by agriculture.
The City of Cape Town continues to demand most of the water in the region, with the local
municipalities of Swartland, Drakenstein and Stellenbosch demanding increasingly more water
from the region.
102
Figure 49: Comparison between urban and agricultural water requirements growth rates
in 2025 and 2040 by municipality
Figure 50: 2025 total annual water requirements by municipality
103
Figure 51: 2040 total annual water requirements by municipality
6.5. Comparative assessment of future water requirements to current economy
The estimation of future water requirements completes the volumetric demand side of the
equation, however, to complete the picture, this forecast demand should be compared to
supply. And then where there is an apparent “supply deficit”, i.e. where supply is unlikely to
meet demand, this deficit should be valued in terms of the economic value that this water would
enable through employment and value-add to the economy. This valuation will provide insight
into where the largest opportunity costs may lie in the regional economy if water supply is not
adequate to meet growth demands. With limited capacity and financial resources, this
calculation will aid regional planners in their prioritisation processes. Thereafter, as the
constraints of water availability are likely to be highly localised, a comparison is made between
the opportunity cost of the supply deficit and the current size of the local economy.
6.5.1. Methodology
The methodology to estimate the “supply deficit” was split into agricultural and urban supply.
The allocation of water to agriculture is capped, nationally and well as within this region (DWS,
2015a). The latest WCWSS reconciliation strategy notes that the total capped allocation to
agriculture is 170 million m3/a, with the data from 2014/15 indicating that 152 million m3/a was
released for agricultural use, but that this was an estimate with a degree of uncertainty as to the
actual usage. There was provision made for agriculture to increase their allocation to 216 million
m3/a, but this is unlikely to happen due to the WCWSS already being over-allocated. It is under-
104
going a Validation and Verification process (refer to Box 1) that will assist in evaluating how
much water is actually being used by agriculture, and furthermore how much of this is lawful.
The recent requirements for agricultural metering will also aid in better establishing actual water
usage. However, regardless of existing usage, allocations are unlikely to increase, especially
when considering the drought in the region, as well as the number of outstanding water use
license applications from Water Services Authorities in the WMA. Therefore any increase in
agricultural demand will result in a supply deficit. This increase in demand may be met with
increased irrigation efficiencies, or by the switching of crops towards less water intense crops.
The estimation of the supply deficit for urban requirements in the region was based on existing
allocations from the WCWSS (DWS, 2015a), industrial WARMS registrations and local supply
sources. The supply augmentation options provided in the WCWSS reconciliation strategy were
not included in the calculation as: a) the purpose of this study is not to provide an alternative
reconciliation strategy, b) with the exception of Voëlvlei’s schemes, all the proposed schemes
are intended for the City of Cape Town. The additional allocation available from Voëlvlei, once
completed, has not been determined and this study could provide data on where the most
economic benefits could be found through this allocation and c) there is still a high degree of
uncertainty as to when/if these schemes will come online.
In some instances, it was difficult to determine the allocation to a certain town or municipality as
it is not contractually set (such as Drakenstein, which receives variable allocations through the
City of Cape Town) or where records were not available for minor towns on the system (2011
demand was utilised in areas that are already using more than allocated). Below are the results
of the allocation by municipality (Table 24):
Table 24: Existing water allocations to urban or industrial users
Municipality Existing allocations
(m3/a) Description
Bergrivier 4,895,370 Allocation by town + WCWSS 2011 demand (Velddrif &
Dwarskersbos) + industrial WARMS
Cape Town 398,700,000 WCWSS allocation and CT dams
Drakenstein 32,435,904 WCWSS + industrial WARMS
Saldanha Bay 11,376,250 WCWSS + Langebaan Rd + Industrial WARMS
Stellenbosch 15,896,904 WCWSS includes estimated 4 mill purchased from CCT (not
fixed) + Industrial WARMS
Swartland 7,242,984 WCWSS allocation for Swartland from Voelvlei + 2011 demand
(Mooreesburg + Koringberg) + Industrial WARMS
Witzenberg 2,245,500 WCWSS allocations for Tulbagh
Total 472,792,912
The existing allocations were then deducted from the predicted demand in 2025 and 2040. The
“business as usual” scenario was utilised for the demand projections with CANEMS2 45 (best
correlated, low-emissions climate model) and historical population figures (from the 2016
community survey), with negative figures representing a potential supply deficit.
Thereafter, the supply deficits were valued according to the per million m3/a employment and
GVA estimates per municipality for urban use (Table 13). Agricultural average value per million
105
m3/a was similarly estimated, with the irrigated water usage linked to the expected crop
requirements per municipality in 2025 and 2040. The valuation of the supply deficit, or the
opportunity cost of insufficient water supply, does not represent the actual cost to the economy
of water constraints. This is impossible to model under the current data constraints and
theoretically complex for allocations so far into the future (there are so many potential shifts in
behaviour, technology and the structure of the economy by 2025 and 2040) and not accurate
when based on average value. However, these opportunity cost estimates were generated to be
compared to the total size of the current economy (2015). The purpose of this exercise was to
highlight where a local economy is likely to feel the most constrained to grow its current
economy due to water scarcity. This is a relative measure of the economic impact of water
availability to a local economy in order to assist regional planners in prioritising support.
6.5.2. Results
The results indicate that, barring any additional allocations or augmentations schemes, the
supply deficit is most keenly going to be felt in the agricultural sector in 2025, with the supply
deficit for urban users only 13% of the total deficit. However this picture changes dramatically by
2040 with an almost even split between the supply deficit of the urban and agricultural
requirements.
Table 25: Urban and agricultural supply deficit in 2025 and 2040 m3/a
Municipality 2025 Urban deficit 27compared to
existing allocation
m3/a
2025 Change in
irrigated
agriculture water
requirements m3/a
2040 Urban deficit
compared to
existing allocation
m3/a
2040 Change in
irrigated
agriculture water
requirements m3/a
Bergrivier 938,982 9,447,272 2,640,940 12,236,017
Cape Town 5,066,997 11,506,373 98,693,543 14,564,853
Drakenstein (2,531,425) 36,656,770 10,843,741 47,092,268
Saldanha Bay 8,225,733 217,849 16,581,091 196,891
Stellenbosch 2,862,026 29,925,972 10,679,324 38,024,375
Swartland 4,866,238 29,564,747 12,464,041 37,444,102
Witzenberg (1,461,110) 4,498,203 (1,550,144) 5,711,880
Total 17,967,442 121,817,185 150,352,536 155,270,386
However, it is important to note how the Saldanha Bay and Swartland municipalities, which are
small users on the WCWSS, have relatively large urban deficits in comparison to their existing
allocations (Table 24). This is because they are already using more than their allocation, and
they are both rapidly growing areas. Cape Town is expected to have the largest urban deficit in
2040, and it therefore stands to reason that most of the schemes planned for the region would
supply the metro. However, given the combined total of all the other urban area deficits in 2040,
the WMA will require more schemes built for the other municipalities in order to meet demand.
The increase in irrigation requirements is substantial with no clear avenue for additional supply
for these users. Competition between urban and agricultural users is to be expected, particularly
27 Positive numbers represent a deficit between expected demand and current allocation (supply), with
negative numbers representing a surplus
106
from riparian farmers who can abstract from the Berg River when releases are made upstream,
possibly destined for downstream dams that supply urban populations. This kind of conflict is
being experienced in the current drought. These results should motivate for further support for
farmers to increase water efficiency and consider the switching to less water intensive and
higher value crops.
Appendix 11 summarises the results on the opportunity costs from water constraints in 2025
and 2040, in terms of the value that could be generated from water availability in 2015 Rands
and employment. Some of the key insights from this analysis are outlined below.
Irrigated agriculture water requirements increase most substantially in 2025 in comparison to
urban requirements, but are still valued lower due to the much higher value created by urban
economies. This can be clearly seen in the 2025 graph below:
Figure 52: 2025 Value of water supply deficit in 2015 R millions
Even when employment is considered (agriculture is an important employer in rural economies),
the comparison still largely values urban water usage, but this varies by municipality.
107
Figure 53: 2025 Value of water supply deficit in employment
When reviewing the results for 2040, Cape Town’s urban requirements outsize the rest of the
municipalities substantially in both rand value and employment impacts.
Figure 54: 2040 Value of water supply deficit in 2015 R millions
108
Figure 55: 2040 Value of water supply deficit in employment
The summary of the total costs per municipality in 2025 and 2040 is included below. As seen in
the graphs above, by 2040, Cape Town’s water requirement deficit and the opportunity cost of
not having this water available is incredibly important for the region overall. However, in 2025,
similar opportunity costs are seen (either in R terms or employment) to the City of Cape Town in
Saldanha Bay, Swartland and Stellenbosch. This is because the City currently has allocation
available that it can still grow into, while Saldanha Bay and Swartland are already using more
than allocated.
Table 26: Water supply deficit opportunity costs in 2025 and 2040
Municipality 2025 total value of
water requirements
deficit in 2015 R
millions
2025 total value of
water requirements
deficit in 2015
employment
2040 total value of
water requirements
deficit in 2015 R
millions
2040 total value of
water requirements
deficit in 2015
employment
Bergrivier 1,200 6,540 3,143 16,229
Cape Town 5,307 23,817 100,614 403,877
Drakenstein 473 8,325 10,496 57,218
Saldanha Bay 4,326 22,152 8,715 44,640
Stellenbosch 2,931 20,884 10,303 56,602
Swartland 5,158 29,771 12,847 68,091
Witzenberg 69 857 89 1,088
Total 19,467 112,347 146,206 647,746
The results from the analysis above, justifies the focus for WCWSS supply augmentation on the
City of Cape Town, as the largest driver of the economy in the region. In order to understand
the impact that these opportunity costs may have on the local economy, and therefore their
109
ability to withstand or absorb them, is shown below through the comparison of the costs in 2025
and 2040 to the current size of the local economy.
Table 27: Comparison of the opportunity cost from a water supply deficit to the current
size of the local economy in 2025 and 2040
Municipality Comparison of
2025 GVA deficit to
2015 total
economy
Comparison of
2025 employment
deficit to 2015 total
employment
Comparison of
2040 GVA deficit to
2015 total
economy
Comparison of
2040 employment
deficit to 2015 total
employment
Bergrivier 33% 30% 86% 74%
Cape Town 2% 2% 29% 29%
Drakenstein 3% 9% 58% 62%
Saldanha Bay 57% 56% 114% 113%
Stellenbosch 22% 32% 77% 87%
Swartland 76% 83% 190% 190%
Witzenberg 1% 2% 1% 2%
Total 5% 7% 36% 38%
In this comparison, it appears that the City of Cape Town is relatively better off in contrast to
some of the other local economies in the region. With particular concern are the West Coast
municipalities of Bergrivier, Saldanha Bay and Swartland. The lack of water availability in these
areas may be a significant constraint to their economic development and ability to generate
jobs, perhaps best illustrated by the below maps of the region:
110
Figure 56: 2025 comparison of GVA deficit to 2015 total economy
Figure 57: 2040 comparison of GVA deficit to 2015 total economy
111
7. Local water and development allocation
prioritisation
7.1. Introduction
As discussed in the previous chapters, the Berg WMA’s readily available water is fully allocated
(termed water-constrained), and there may be a significant time lag until new resources come
online and water availability can increase. It is proposed that allocation in these constrained
situations should be towards those water users or developments that maximise socio-economic
benefits for the water used. Strategically allocating water is relevant at both a regional and local
level.
Saldanha Bay Local Municipality (SBLM), as a downstream user in the Berg WMA, has
recognised the need to foster a decision-making environment which can effectively manage the
municipality’s water service delivery requirements through a systematic and explicit treatment of
trade-offs. The provision of water is challenging in SBLM as the area has experienced
significant population growth in recent times, with supply levels remaining constant. Adding to
these challenges, the expansion of industrial activity in the area could soon result in a situation
where demand for potable water exceeds supply (the municipality is already abstracting more
from the WCWSS than their licence allows). The effective allocation of available water towards
the municipality’s development priorities is thus of paramount importance.
Interviews with local decision-makers in economic development planning and water resources
planning revealed that proposed industrial projects are currently assessed individually and
awarded water allocations essentially on a first-come-first-served basis. However, there was
also an acknowledgement that this approach is not ideal and that tools and information required
to be more strategic would be welcomed. Out of these discussions came the idea of developing
a tool that could assist in the prioritisation of requests for water allocation from industry on the
basis of the development gains to be realised.
Furthermore, GreenCape had gathered data on major proposed developments in the area during
the scoping phase of the project (GreenCape, 2015) and their associated water requirements.
Subsequent to the scoping study, DED&T commissioned the West Coast Industrial Plan (WCIP):
a study to detail all major proposed developments in the area, evaluating all of their major
resource requirements (land, energy, roads, skills requirements etc) (DED&T, 2016). This study
formed a required data baseline for the local prioritisation tool. Maintaining an updated repository
of information on proposed developments, and the water requirement thereof, is an on-going
process that, since the scoping phase, DED&T has recognised as necessary and will need to
continue with.
7.2. Multi-Criteria Decision Analysis (MCDA) and municipal decision-making
In discussions with the SBLM officials, it was apparent that water required from a new
development was the most important criteria that they wished to evaluate in comparison to the
other benefits and costs that a development may bring to the area, but that water demand alone
would not be the only basis for a decision. It was also recognised that decision-making would
likely require the input of multiple individuals with varying and possibly even competing goals.
Cost-Benefit Analysis (CBA) was considered, but given that the approach requires variables to
112
be monetised, it was found to be unsuitable. After some discussion, a Multi-Criteria Decision
Analysis (MCDA) approach was identified as most appropriate for providing general guidance in
the development of the tool. MCDA allows decision-makers to evaluate alternative options
based on a diverse range of criteria, regardless of the units used to measure them. The criteria
can be assigned varying levels of importance or weight, allowing for differing levels of
prioritisation. The result is a set of weighted scores which can be used to evaluate alternatives.
The approach also lends itself particularly well to group decision-making environments
(Geneletti, 2007). Furthermore, the availability of the recently published WCIP was fortuitous,
given that the study details all the key proposed developments in the Saldanha Bay area and
includes an evaluation of their major resource requirements (water, land, energy, transport, etc.)
and some of their likely socio-economic benefits (operational jobs, construction jobs, capital
investment) for the area.
Having gained an understanding of the approach which would be used as well as a potential
data sources, a series of meetings were held with representatives from SBLM. There was an
iterative process of identifying ideal criteria to be used in evaluating industrial projects, and
scrutinising how practical they were primarily based on data availability and/or the ease with
which they could be estimated by managers. This process resulted in a list of five criteria, three
of which could be informed directly from the WCIP dataset and two of which would require
considered input from various municipal managers with adequate knowledge of the proposals
being considered.
A succession of workshops and internal project meetings were also held to improve the integrity
and robustness of the model. Input was thus received from professionals in the private sector,
provincial government representatives and academics all of which informed iterations. The
development of the model was strongly guided by the desire to produce a user-friendly product
to aid decisions and highlighting trade-offs between alternatives in an environment of competing
interests. Ultimately the tool’s utility lies in its ability to enrich the user’s decision-making
process.
7.3. Overview of the tool
This section provides step-by-step guidance on how to use the tool. There are five sections,
outlining the various Microsoft Excel sheets which comprise the tool. A schematic overview is
provided Figure 58, which shows how the processes of alternative selection, criteria selection,
criteria scoring and criteria weighting come together to generate results which can inform
decision-making.
113
Figure 58: Schematic showing the different processes making up the tool
The following sub-sections contain a closer look at each of the above processes using an
instructional style.
7.3.1. Selection of alternatives
The first process has one step – list the alternatives that you wish to consider in the box. A
maximum of 30 alternatives can be entered and compared using the tool.
Step 1: List alternatives in the blue box below
Alternatives
Figure 59: Selection of alternatives
7.3.2. Criteria selection and metrics
This process will establish which criteria the alternatives are to be measured by, how they are to
be measured, and briefly consider the users’ preferences towards the criteria. This entails five
steps.
Selection of alternatives
Criteria selection
Criteria scoring
Criteria weighting
Results
114
7.3.2.1. Step 1
Enter the range of criteria which are to be
considered. Begin with the most important criterion.
This will be the reference criterion, and will be used
later when determining the relative importance of all
the other criteria. Beyond this first criterion, the
ordering of the other criteria is unimportant.
Step 1: In the blue column: Enter the criteria by which you wish to evaluate the alternatives placing your most important or reference criteria at the top of the list
Step 2: In the green column, enter the metric used to measure the criteria by (eg. kl/day where data are available or 'rating' where alternatives can be rated individually without data)
Step 3: In the red column: Specify whether you would prefer this criteria to have a higher value or a lower value (to inform the next step)
Step 4: In the yellow column: Enter the worst case acceptable value which an alternative can have for the criteria and still be considered (for ratings we suggest using 0)
Step 5: In the grey column: Enter the best case acceptable value for the criteria (for ratings we suggest using 10)
Rank Criterion MetricBetter if lower or
higher?Worst case acceptable value Best case acceptable value
1
2
3
4
5
6
7
8
9
10
Notes and clarifications (use this box to take any notes that you might need):
1
2
3
Rank Criterion
1
2
3
Criteria (singular - criterion) are the
attributes selected to be used in
evaluating an alternative. For
example, one might evaluate a car
based on its price, its rate of fuel
consumption, or perhaps by engine
performance.
Figure 60: Criteria definition
115
7.3.2.2. Step 2
Enter the metrics by which the various criteria will be measured. Where data are available, these
will simply be the units in which the data are represented. Where criteria are to be rated by the
user, the metric could simply be ‘rating’.
Metric
Figure 61: Criteria Metrics
116
7.3.2.3. Step 3
Next, decide whether it would be preferable for the value of each criterion to be lower or higher.
In the red column, enter either ‘lower’ or ‘higher’. For example, one might favour a project with a
low potable water use, but when it comes to a criterion such as operational jobs, a higher figure
would probably be preferred.
Better if lower or
higher?
Figure 62: Criteria positive or negative
relationship
117
7.3.2.4. Step 4
During this step, decide what the worst case acceptable value would be for each criterion.
Where the preference is for the criterion to be higher, this value would be relatively low and
usually set to “0”. Where criteria are preferred to be lower, this value can be set at the cut-off
point beyond which alternatives would not be considered at all.
Worst case acceptable
value
Figure 63: Criteria worst case acceptable value
118
7.3.2.5. Step 5
Decide on the best-case acceptable value for each criterion. This figure should be arrived at by
carefully considering what the best-case value is that one could expect for each criterion. Where
the goal is to have as low a value as possible this could simply be “0”, but where the goal is to
have a large amount (ie. where it is better for the criterion to be higher), care must be taken to
ensure that none of the alternatives end up with a higher number than what is decided on for this
value. At the same time, the number should not be unreasonably high, and it should at least be
conceivable that an alternative could have a value this high for each particular criterion. If the
criterion is a rating-based and better when higher, the most straightforward approach would be to
set this value to “100”.
Best case acceptable
value
Figure 64: Criteria best case acceptable value
119
7.3.3. Criteria scoring
The criteria scoring process must be
completed for each of the criteria selected in
Section 7.3.2.1. For each criterion, there are
4 steps guiding the user through the
formulation of a value function and the
subsequent scoring of each alternative.
7.3.3.1. Step 1
This step only applies if there are no data on the criterion being considered, and each project
will thus need to be rated based on the knowledge held by the user. Where data do exist,
proceed directly to Step 2. Where projects will need to be rated, use the blue box to develop a
legend outlining information for rating alternatives, indicating what different ratings would mean
in terms of project characteristics.
Criterion:
Step 1: If there are data on this criterion, go to step 2. If not, use the blue box to clarify what the different ratings, between 0 and 10, mean.
Step 2: In the green column, create a scoring scale by entering the data or rating which would correspond to each of the scores out of 100.
Step 3: In the red column, place an x in the box next to the shape which looks most like the value function created in the 'Value Function' graph.
Step 4: In the yellow column, if data exists, enter it here, otherwise rate the alternative using the informants developed in Step 1.
Step 5: Use the data in the yellow column, along with the scoring scale and value function, to allocate scores to each of the alternatives.
0 0
10
20
30
40
50
60
70
80
90
0 100
Legend/Informants
0 =
5 =
10 =
Linear
Exponential
Logarithmic
Quadratic
0 Score
Scoring Scale
0
Alternative Score
0
10
20
30
40
50
60
70
80
90
100
0 0 0 0 0 1 1 1 1 1 1
Value Function
or
or
Figure 65: Criterion legend
0 = 5 = 10 =
Legend/Informants
The goal of formulating a value function is
to be able to express (in mathematical and
graphical formats) the value which the user
attaches to a criterion and how this value
changes according to the range of quantities
in which it could potentially be experienced.
120
7.3.3.2. Step 2
Consider the range of values which fall in between the worst case acceptable value and the
best case acceptable value, and how these are to be distributed. In the green table, shown
below, the worst-case acceptable value is automatically allocated a score of 0, and the best-
case value is automatically scored at 100. Decide which values would be allocated a score of
10, 20, 30… up to 90. Instead of performing this exercise in a linear way, enter the values which
you feel the most confident about first, such as the mid-way score of 50. Alternatively, perform
the thought exercise of considering what a particular value would score for a number of different
values and check the consistency of these scores with the overall value function.
As values are allocated to the various scores, the value function will materialise automatically in
the graph next to the table (i.e. the tool has been programmed to generate value functions). You
can then use this function as a guide to recalibrate how the different values relate to the scores.
It is helpful to experiment with different numbers at this stage, revisiting and changing values
until the graph shows a value function which you believe accurately reflects your preferences.
Scoring Scale
0 Score
0 0
10
20
30
Figure 66: Criterion scoring scale
121
7.3.3.3. Step 3
Once values have been allocated to the different scores, the value function will have taken on a
definite shape. Compare this shape to the pre-defined shapes in the table below and check the
box next to the shape which most resembles the value function arrived at.
7.3.3.4. Step 4
Once the value function has been established and the shape identified, the next step is to consider
the alternatives being evaluated. In the final box on this sheet, enter data on the individual
alternatives in the yellow column. For some criteria, data will already exist. For example, for
operational jobs one will need to enter the number of operational jobs which each alternative is
likely to create. For other criteria, a judgement may be required. For example, if the criterion is
measured as a score from 0 to 100, manually rate each project on a scale between 0 and 100.
Figure 67: Value function shapes
122
7.3.3.5. Step 5
Using the data entered in the yellow column, along with the Scoring Scale and Value Function as
guiding tools, enter a score for each alternative in the grey column.
The procedure outlined in Steps 1 to 5 of this section will need to be completed several times,
depending on the number of criteria selected on the ‘Criteria’ page. For example, if alternatives
are being evaluated based on 3 criteria, then the sheets titled ‘Crit 1’, ‘Crit 2’ and ‘Crit 3’ all need
to be filled out.
7.3.4. Criteria weighting
The final sheet to consider is the ‘Weighting’ sheet. The purpose of this sheet is to determine the
relative importance of the different criteria. Note that there are two ways this can be done:
1. Answer the trade-off questions in the blue boxes (this is the more theoretically correct
manner to calculate weights)
Alternative Score
Figure 68: Alternative scoring
123
2. Enter weights manually in the red boxes (can also work well particularly when trade-offs
are understood)
The instructions below will proceed on the basis that the user wishes to try the first approach and,
failing a satisfactory result, move to the second approach.
7.3.4.1. Step 1
On the left-most part of the sheet, the user will find questions determined using the answers given
in previous parts of the exercise. The number of questions will be one less than the number of
criteria specified. For example, if three criteria were entered previously, there will be two
questions. The questions use the first criterion entered as a reference criterion, and are set up in
a way as to determine, for any given alternative, how much of the reference criterion one would
be willing to trade-off against the full amount of each of the other criteria. On the right side is a
box with blue-shaded cells. These cells are where the questions must be answered. Provision is
made for two sensitivity tests. That is, one has the option to enter one answer in the
‘selected/favoured’ column, as well as two additional answers in the ‘sensitivity 1’ and ‘sensitivity
2’ columns.
124
7.3.4.2. Step 2
Once the blue cells have been populated, the tool will take the answers given, combined with the
value functions determined previously, and calculate the inferred weighting for each criterion. This
weighting should reflect the relative importance of each of the criteria. A higher weighting
suggests that a criterion is more important. Altogether the weights should add up to 100.
Step 2 simply entails checking that the weightings generated are an accurate reflection of the
importance which each of the criteria has for the decision at hand.
Note: There are 2 ways to complete this process:1. Answer the trade-off questions in the blue boxes (generally thought of as more theoretically correct)2. Enter weights manually in the red boxes (can also work well particularly when trade-offs are understood)
Step 1: In the blue boxes: Answer the trade-off questions displayed below
Step 2: In the green table: Check that the implied weights generated seem reasonable
Step 3 (Optional): If the implied weights do not make sense, try going through the blue trade-off questions again
1 What level of Criteria 1 x units of Criteria 2 ?
2 What level of Criteria 1 y units of Criteria 3 ?
3 What level of Criteria 1 z units of Criteria 4 ?
Base case Sensitivity 1 Sensitivity 2
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
0.0 0.0 0.0
Weighting
Total
Criterion
Criterion 1
Criterion 2
Criterion 3
0
0
0
0
0
0
0
Step 4 (Optional): If the implied weights still do not make sense, they can be manually entered in the red table (place an "x" in the box)
Note: The implied weights are generated using a
combination of the value functions calculated on
previous sheets and the answers given to the
trade-off questions above
Note: Sensitivity 1 and Sensitivity 2 are optional
and can be used to compare the results of
different answers
Base case Sensitivity 1 Sensitivity 2
would you be willing to allow for a project which also involves
would you be willing to allow for a project which also involves
would you be willing to allow for a project which also involves
Implied Weight
Criterion
Criterion 1
Criterion 2
Criterion 3
0
0
0
0
0
0
If you are not satisfied with
the outomes of the question-
based weightings, place an
"X" in the box below and do
manual weightings in the
red cells
Manual Weight
Base case Sensitivity 1 Sensitivity 2
Total 0.0
0
0.0 0.0
Figure 69: Weighting questions
1 What level of Criteria 1 x units of Criteria 2 ?
2 What level of Criteria 1 y units of Criteria 3 ?
3 What level of Criteria 1 z units of Criteria 4 ?
would you be willing to allow for a project which also involves
would you be willing to allow for a project which also involves
would you be willing to allow for a project which also involves
Base case Sensitivity 1 Sensitivity 2
125
7.3.4.3. Step 3
If the weightings generated do not seem reasonable, it is recommended that the answers provided
to the questions be reconsidered and adjusted if necessary. This process can be repeated until
you are comfortable with the weightings generated.
7.3.4.4. Step 4
If, after reconsidering the answers provided, the implied weights generated still do not seem
correct, the user may consider the manual weighting option. To select the manual weighting
Note: There are 2 ways to complete this process:1. Answer the trade-off questions in the blue boxes (generally thought of as more theoretically correct)2. Enter weights manually in the red boxes (can also work well particularly when trade-offs are understood)
Step 1: In the blue boxes: Answer the trade-off questions displayed below
Step 2: In the green table: Check that the implied weights generated seem reasonable
Step 3 (Optional): If the implied weights do not make sense, try going through the blue trade-off questions again
1 What level of Criteria 1 x units of Criteria 2 ?
2 What level of Criteria 1 y units of Criteria 3 ?
3 What level of Criteria 1 z units of Criteria 4 ?
Base case Sensitivity 1 Sensitivity 2
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
0.0 0.0 0.0
Weighting
Total
Criterion
Criterion 1
Criterion 2
Criterion 3
0
0
0
0
0
0
0
Step 4 (Optional): If the implied weights still do not make sense, they can be manually entered in the red table (place an "x" in the box)
Note: The implied weights are generated using a
combination of the value functions calculated on
previous sheets and the answers given to the
trade-off questions above
Note: Sensitivity 1 and Sensitivity 2 are optional
and can be used to compare the results of
different answers
Base case Sensitivity 1 Sensitivity 2
would you be willing to allow for a project which also involves
would you be willing to allow for a project which also involves
would you be willing to allow for a project which also involves
Implied Weight
Criterion
Criterion 1
Criterion 2
Criterion 3
0
0
0
0
0
0
If you are not satisfied with
the outomes of the question-
based weightings, place an
"X" in the box below and do
manual weightings in the
red cells
Manual Weight
Base case Sensitivity 1 Sensitivity 2
Total 0.0
0
0.0 0.0
Figure 70: Implied weights
Base case Sensitivity 1 Sensitivity 2
1
2
3
Criterion
Criterion 1
Criterion 2
Criterion 3
Implied Weight
126
option, place an ‘x’ in the box between the two weighting tables and proceed to enter the manual
weights in the red columns as shown below.
7.3.5. Generated results
The results sheet will automatically display the alternatives and their scores under the ‘base case’
weighting, the ‘sensitivity 1’ weighting and the ‘sensitivity 2’ weighting respectively. Here it is
useful to consider how well each alternative performed under the three different weighting
combinations. For example, one alternative might achieve a high score under the ‘base case’
weighting combination and a lower score under the ‘sensitivity 1’ weighting combination, based
Note: There are 2 ways to complete this process:1. Answer the trade-off questions in the blue boxes (generally thought of as more theoretically correct)2. Enter weights manually in the red boxes (can also work well particularly when trade-offs are understood)
Step 1: In the blue boxes: Answer the trade-off questions displayed below
Step 2: In the green table: Check that the implied weights generated seem reasonable
Step 3 (Optional): If the implied weights do not make sense, try going through the blue trade-off questions again
1 What level of Criteria 1 x units of Criteria 2 ?
2 What level of Criteria 1 y units of Criteria 3 ?
3 What level of Criteria 1 z units of Criteria 4 ?
Base case Sensitivity 1 Sensitivity 2
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
0.0 0.0 0.0
Weighting
Total
Criterion
Criterion 1
Criterion 2
Criterion 3
0
0
0
0
0
0
0
Step 4 (Optional): If the implied weights still do not make sense, they can be manually entered in the red table (place an "x" in the box)
Note: The implied weights are generated using a
combination of the value functions calculated on
previous sheets and the answers given to the
trade-off questions above
Note: Sensitivity 1 and Sensitivity 2 are optional
and can be used to compare the results of
different answers
Base case Sensitivity 1 Sensitivity 2
would you be willing to allow for a project which also involves
would you be willing to allow for a project which also involves
would you be willing to allow for a project which also involves
Implied Weight
x
Criterion
Criterion 1
Criterion 2
Criterion 3
0
0
0
0
0
0
If you are not satisfied with
the outomes of the question-
based weightings, place an
"X" in the box below and do
manual weightings in the
red cells
Manual Weight
Base case Sensitivity 1 Sensitivity 2
Total 0.0
0
0.0 0.0
Figure 71: Manual weights
1
2
3
Manual Weight
Base case Sensitivity 1 Sensitivity 2Criterion
Criterion 1
Criterion 2
Criterion 3
x
127
on the different emphases which each of the weighting combinations place on the various criteria.
The layout is shown below.
Figure 72: Alternative results
At this point it is useful to consider the different trade-offs involved between criteria and whether
the weighted score provides a fair representation of your overall preference for each of the
alternatives.
7.4. Application of tool in Saldanha Bay
This section demonstrates the workings of the tool by showing an example – the ranking of a
group of industrial projects proposed in the Saldanha Bay area. Alternatives were selected purely
for illustrative purposes to reflect a broad selection of the types of developments currently
proposed in the area. The findings are the result of three meetings held with the SBLM and reflect
the opinions of the municipal officials present. During these meetings, various representatives
provided input regarding the selection and prioritisation of criteria, construction of value functions,
rating of alternatives and weighting of the various criteria.
7.4.1. Alternatives
A subset of projects was chosen from the West Coast Industrial Plan (WCIP). As the exercise
was one to demonstrate the tool and familiarise the municipal officials with the tool, the projects
were chosen purely because they illustrate typical projects in different sectors and as adequate
data was available for them.
Figure 73: WCIP alternatives
Base case Sensitivity 1 Sensitivity 2
Weighted Score
Alternative
Step 1: List alternatives in the blue box below
Alternatives
IDZ
Afrisam - Phase 1
Elandsfontein 1
ArcelorMittal
Mineral beneficiation plant
TNPA Rig repair at Berth 205
128
7.4.2. Criteria selection and metrics
The selection of criteria was guided by a consideration of municipal priorities, as well as by the
recognition of constraints in data availability.
7.4.2.1. Step 1
Five criteria were identified by municipal representatives as important and feasible for inclusion
in the analysis. The criterion deemed to be most important was potable water use, and it was thus
chosen as the reference criterion. The other criteria selected by the SBLM representatives were
non-potable water use, operational jobs, potential for local jobs and benefits, and environmental
impacts, as shown below.
7.4.2.2. Step 2
The first two criteria, potable and non-potable water use, were measured in ‘kilolitres per day’.
Operational jobs are measured as a discrete number – ‘jobs’. Both potential for local jobs and
Step 1: In the blue column: Enter the criteria by which you wish to evaluate the alternatives placing your most important or reference criteria at the top of the list
Step 2: In the green column, enter the metric used to measure the criteria by (eg. kl/day where data are available or 'rating' where alternatives can be rated individually without data)
Step 3: In the red column: Specify whether you would prefer this criteria to have a higher value or a lower value (to inform the next step)
Step 4: In the yellow column: Enter the worst case acceptable value which an alternative can have for the criteria and still be considered (for ratings we suggest using 0)
Step 5: In the grey column: Enter the best case acceptable value for the criteria (for ratings we suggest using 10)
Rank Criterion MetricBetter if lower or
higher?Worst case acceptable value Best case acceptable value
1 Potable water use Kl/day lower 5000 0
2 Non-potable water use Kl/day higher 0 1000
3 Operational job creation jobs higher 0 3000
4 Potential for local jobs and benefits rating higher 0 10
5 Environmental impacts rating higher 0 10
6 Criterion 6
7 Criterion 7
8 Criterion 8
9 Criterion 9
10 Criterion 10
Notes and clarifications:
1
2
3
Potential for local jobs and benefits will also partially reflect indirect strain on municipal services (i.e. projects that employ more locals result in less in-migration and therfore less
risk of additional strain on services).
Figure 74: Selected criteria for SBLM
evaluation
Rank Criterion
1 Potable water use
2 Non-potable water use
3 Operational job creation
4 Potential for local jobs and benefits
5 Environmental impacts
129
benefits and environmental impacts are criteria for which data did not exist. Alternatives were
therefore evaluated on a scale from 0 to 10, using a ‘rating’.
7.4.2.3. Step 3
Potable water use was found to be a negative criterion. Thus, a project which used less potable
water was preferred to a project which used more. It was therefore better for the value to be lower.
For all other criteria, a higher value was found to be preferable.
Figure 75: SBLM
criteria metrics
Metric
Kl/day
Kl/day
jobs
rating
rating
130
7.4.2.4. Step 4 and Step 5
For potable water use, the worst case acceptable value was arrived at by considering projects
which have in the past been denied water allocation rights because their demand was too high.
This experience allowed the decision makers to confidently assert that the worst case acceptable
value was 5 000 kl/day. Non-potable water use had a positive association, and thus the worst
case acceptable value was deemed to be 0 kl/day. Operational job creation was also considered
a positive criterion, and the worst case acceptable value was thus set at 0 jobs. The worst-case
rating for potential for local jobs and benefits was set to 0, as was the worst-case rating for
environmental impacts.
The best-case value for potable water use was found to be 0 kl/day. In the case of non-potable
water use, municipal representatives considered 1 000 kl/day to be the best case acceptable
value that they could hope for. The best case acceptable value for operational jobs was set to 3
Figure 76: Definition
of criteria as positive
or negative
relationship
Better if lower or
higher?
lower
higher
higher
higher
higher
131
000 jobs. For both potential for local jobs and benefits and environmental impacts, the best case
acceptable rating was set at 10.
7.4.3. Criteria scoring
This following section will present results from the processes used to determine the project scores
for each of the criteria through the construction of individual value functions.
7.4.3.1. Potable water use
Here the municipal representatives were required to determine the shape of the value curve for
potable water use. Given that the worst case acceptable value (5 000 kl/day) and the best case
acceptable value (0 kl/day) had already been entered, the municipal representatives proceeded
to consider a relatively noteworthy value in between, asking themselves “if a project were to use
1 000 kl/day of potable water, what sort of score would they get?”. After some debate, it was
decided that such a project would score 60 points. The value function on the right of the table now
began to display a graph, which could be used to calibrate the answers as they went. The
Worst case acceptable value Best case acceptable value
5000 0
0 1000
0 3000
0 10
0 10
Figure 77: Criteria best and worst case values
132
Figure 78: Potable water use value function graph
municipal representatives then asked themselves what a project would score if it used 4 000
kl/day of potable water. The answer was found to be 10 points. The value function had now begun
to take shape and the municipal representatives could rely on it more to infer what their answers
would be for projects scoring between 10 and 60, and between 60 and 100. Based on the
graphical representation of the created value function, the municipal representatives could infer
that the function looked most like an exponential one, and the appropriate box was selected.
Having entered project potable water use data in the yellow column, the municipal
representatives went on to generate the scores shown below. It can be seen that TNPA Rig
repair at Berth 205 scored the maximum of 100 points, having a potable water use of 0 kl/day.
Afrisam – Phase 1’s potable water use was also very close to 0, and so the project was given
99 points. The project which scored the lowest was the IDZ, using 1 453 kl/day and thus scoring
46 points.
Figure 79: Potable water use scores
Figure 81: Potable water use shape
7.4.3.2. Non-potable water use
Establishing the value function for non-potable water use was relatively straightforward. The
municipal representatives had already established that the best case acceptable use of non-
5000 0
4000 10
2750 20
2250 30
1750 40
1350 50
1000 60
800 70
500 80
200 90
0 100
Kl/day Score
Scoring Scale
x
Linear
Exponential
Logarithmic
Quadratic
or
or
IDZ 1453 46
Afrisam - Phase 1 8 99
Elandsfontein 1 160 94
ArcelorMittal 411 82
Mineral beneficiation plant 1227 53
TNPA Rig repair at Berth 205 0 100
Alternative Kl/day Score
Figure 80: Potable water use project alternative values and
scores
133
IDZ 0 0
Afrisam - Phase 1 100 10
Elandsfontein 1 3040 100
ArcelorMittal 0 0
Mineral beneficiation plant 0 0
TNPA Rig repair at Berth 205 0 0
ScoreAlternative Kl/day
Figure 84: Non-potable water use project alternative
values and scores
potable water was 1 000 kl/day. Any project using more than or equal to this value would thus get
100 points. The municipal representatives felt that their preferences for non-potable water use
were linear, and that projects using 100 kl/day should receive 10 points, 200 kl/day - 20 points,
and so forth.
The linear function was thus selected with an ‘x’ and the non-potable water use data was entered
for each project. The projects were then scored, as shown below. Most of the projects did not
report any non-potable water use requirements and thus scored 0. According to the WCIP,
Elandsfontein 1 had a requirement of 3 040 kl/day. Because this is greater than the best case
acceptable 1 000 kl/day, this project scored the full 100 points.
Figure 83: Non-potable water use scores
Figure 85: Non-potable water use shape
7.4.3.3. Operational job creation
In establishing the value function for operational job creation, municipal representatives began by
considering how many jobs a project would have to generate to score 50 points. The answer was
thought to be 300 jobs. Given that the best-case number of jobs was thought to be 3 000, the
value function immediately showed a somewhat logarithmic shape. Projects generating anywhere
0 0
100 10
200 20
300 30
400 40
500 50
600 60
700 70
800 80
900 90
1000 100
Kl/day Score
Scoring Scale
Quadratic
x
Exponential
Logarithmic
Linear or
or
Figure 82: Non-potable water use value function
graph
134
Quadratic
Exponential
Logarithmic x
Linear or
or
IDZ 2602 92
Afrisam - Phase 1 113 22
Elandsfontein 1 284 48
ArcelorMittal 45 8
Mineral beneficiation plant 181 37
TNPA Rig repair at Berth 205 2800 97
ScoreAlternative jobs
Figure 86: Operational jobs value function graph
Figure 89: Operational jobs
shape
Figure 88: Operational jobs alternatives values and
scores
between 0 and 10% of the best-case number of jobs would score between 0 and 50% of the
maximum score. Therefore, 300 can be thought of as a threshold figure in this case, and the
graphical representation of the value function shown below shows that the function is almost linear
on either side of 300.
Overall, the function was thought to be logarithmic and, having entered the data on each of the
projects the municipal representatives went on to generate scores. ArcelorMittal scored the lowest
(8) as it would generate the fewest jobs (45). TNPA Rig repair at Berth 205 would generate the
highest number of jobs (2 800) and thus scored 97 points.
Figure 87: Operational jobs scores
0 0
50 10
100 20
150 30
200 40
300 50
800 60
1300 70
1800 80
2500 90
3000 100
jobs Score
Scoring Scale
135
Figure 90: Potential for local jobs value function graph
7.4.3.4. Potential for local jobs and benefits
The potential for local jobs and benefits criterion did not have any data and was thus chosen to
be a rating-based criterion. The following guidance was developed to inform the project ratings:
Legends/Informants
0 = not compatible with socio-economic development goals with very limited benefit for local
economy (all employees are from elsewhere, all expenditure is directed elsewhere)
5 = moderately compatible with socio-economic development goals with moderate impact on local
economy (50/50 split between employees that are local and from elsewhere, expenditure partially
sourced locally, partially from elsewhere)
10 = highly compatible with socio-economic development goals with maximal impact on local
economy (75/25 or higher split between employees that are local and from elsewhere, high level of
local expenditure)
To keep things simple, the value function was made linear, with the ratings varying proportionately
with score (1 = 10, 2 = 20, 3 = 30, etc.). The rating box was updated accordingly and an ‘x’ placed
next to the linear curve.
The municipal representatives were then asked to rate the projects. Afrisam – Phase 1 and
Elandsfontein were both predicted to be ‘highly compatible with socio-economic development
goals’ in the local area and were thus rated 10 each, translating into scores of 100 each. The
TNPA Rig repair at Berth 205 was thought to be the least compatible with local development
goals, receiving a rating of 5 (score of 50).
Figure 91: Potential for local jobs scores
0 0
1 10
2 20
3 30
4 40
5 50
6 60
7 70
8 80
9 90
10 100
rating Score
Scoring Scale
136
Figure 93: Potential for local jobs shape
7.4.3.5. Environmental Impacts
It was decided that a rating would also work best to capture the complex array of factors used to
measure a project’s environmental impacts. The following guideline was developed to inform the
rating process:
Legend/Informants
0 = very high impacts on any of the key environmental factors (biodiversity loss, air pollution or water
pollution) regardless of mitigation measures
1 to 3 = high impacts on any of the key environmental factors, 1 = where mitigation measures have
been included but are at risk of not being sufficiently implemented (technical, regulatory or cost
constraints) and 3 = where mitigation measures are highly likely to be sufficiently implemented
4 to 6 = medium impacts on any of the key environmental factors, 4 = where mitigation measures
have been included but are at risk of not being sufficiently implemented and 6 = where mitigation
measures are highly likely to be sufficiently implemented
7 to 9 = low impacts on any of the key environmental factors, 7 = where mitigation measures have
been included but are at risk of not being sufficiently implemented and 9 = where mitigation
measures are highly likely to be sufficiently implemented
10 = no environmental impacts expected
The value function for environmental impacts was also made linear. The rating box was updated
accordingly and the relevant curve was selected with an ‘x’.
All of the alternative projects being considered were found to have either medium or high impacts
on the environment. Afrisam – Phase 1 and the Mineral beneficiation plant were found to have
Quadratic
x
Exponential
Logarithmic
Linear or
or
IDZ 7 70
Afrisam - Phase 1 10 100
Elandsfontein 1 10 100
ArcelorMittal 6 60
Mineral beneficiation plant 6 60
TNPA Rig repair at Berth 205 5 50
Alternative rating Score
Figure 92: Potential for local jobs alternatives values
and scores
137
Figure 94: Environmental impact value function graph
the highest impacts on the environment and thus scored 30 points each. The rest of the projects
scored either 50 or 60, as seen below.
Figure 95: Environmental impact scores
Figure 97: Environmental impact shape
7.4.4. Criteria Weighting
The final sheet to consider was the ‘Weighting’ sheet. The purpose of this sheet was to determine
the relative importance of the different criteria. The municipal managers opted to go for the first
of the two approaches, using the questions to generated implied weights.
0 0
1 10
2 20
3 30
4 40
5 50
6 60
7 70
8 80
9 90
10 100
rating Score
Scoring Scale
Quadratic
x
Exponential
Logarithmic
Linear or
orIDZ 6 60
Afrisam - Phase 1 3 30
Elandsfontein 1 5 50
ArcelorMittal 6 60
Mineral beneficiation plant 3 30
TNPA Rig repair at Berth 205 6 60
Alternative rating Score
Figure 96: Environmental impact values and scores
138
7.4.4.1. Step 1
The municipal stakeholders proceeded to answer the questions generated by the tool,
considering the trade-offs implied in each. The questions were answered as shown below. It was
decided that one set of answers would be sufficient, and so only the base-case questions were
answered.
Note: There are 2 ways to complete this process:1. Answer the trade-off questions in the blue boxes (generally thought of as more theoretically correct)2. Enter weights manually in the red boxes (can also work well particularly when trade-offs are understood)
Step 1: In the blue boxes: Answer the trade-off questions displayed below
Step 2: In the green table: Check that the implied weights generated seem reasonable
Step 3 (Optional): If the implied weights do not make sense, try going through the blue trade-off questions again
1 What level of Potable water use 1000 Kl/day of Non-potable water use ? Answer: 1000 Kl/day Kl/day Kl/day
2 What level of Potable water use 3000 jobs of Operational job creation ? Answer: 2000 Kl/day Kl/day Kl/day
3 What level of Potable water use 10 rating of Potential for local jobs and benefits ? Answer: 2000 Kl/day Kl/day Kl/day
4 What level of Potable water use 10 rating of Environmental impacts ? Answer: 1000 Kl/day Kl/day Kl/day
Base case Sensitivity 1 Sensitivity 2
1 31.9 1
2 12.8 2
3 21.3 3
4 21.3 4
5 12.8 5
6 0.0 6
7 0.0 7
8 0.0 8
9 0.0 9
10 0.0 10
100.0 0.0 0.0 Total 0.0
Criterion 10
0.0 0.0
Manual Weight
Base case Sensitivity 1 Sensitivity 2Criterion
Potable water use
Non-potable water use
Operational job creation
Potential for local jobs and benefits
Environmental impacts
Criterion 6
Criterion 7
Criterion 8
Criterion 9
If you are not satisfied with
the outomes of the question-
based weightings, place an
"X" in the box below and do
manual weightings in the
red cells
Note: The implied weights are generated using a
combination of the value functions calculated on
previous sheets and the answers given to the
trade-off questions above
Note: Sensitivity 1 and Sensitivity 2 are optional
and can be used to compare the results of
different answers
Base case Sensitivity 1 Sensitivity 2
would you be willing to allow for a project which also involves
would you be willing to allow for a project which also involves
would you be willing to allow for a project which also involves
would you be willing to allow for a project which also involves
Implied Weight
Weighting
Total
Criterion
Potable water use
Non-potable water use
Operational job creation
Potential for local jobs and benefits
Environmental impacts
Criterion 6
Criterion 7
Criterion 8
Criterion 9
Criterion 10
Step 4 (Optional): If the implied weights still do not make sense, they can be manually entered in the red table (place an "x" in the box)
Figure 98: Weighting question responses
1 What level of Potable water use 1000 Kl/day of Non-potable water use ?
2 What level of Potable water use 3000 jobs of Operational job creation ?
3 What level of Potable water use 10 rating of Potential for local jobs and benefits ?
4 What level of Potable water use 10 rating of Environmental impacts ?
would you be willing to allow for a project which also involves
would you be willing to allow for a project which also involves
would you be willing to allow for a project which also involves
would you be willing to allow for a project which also involves
Answer: 1000 Kl/day Kl/day Kl/day
Answer: 2000 Kl/day Kl/day Kl/day
Answer: 2000 Kl/day Kl/day Kl/day
Answer: 1000 Kl/day Kl/day Kl/day
139
7.4.4.2. Step 2
Once the questions had been answered, the tool generated the inferred weighting for each
criterion. The final task was to verify that the weightings generated were an accurate reflection of
the importance which each of the criteria have for the decision at hand. The weightings generated
were found to be satisfactory to the municipal representatives. They are shown below.
7.4.5. Generated Results
The automatically generated results are outlined below. The best performing project was
Elandsfontein 1 (81) given the high scores achieved for most of the criteria, particularly for potable
Criterion
Implied
Weight
Base
case
1 Potable water use 31.9
2 Non-potable water use 12.8
3 Operational job creation 21.3
4 Potential for local jobs and benefits 21.3
5 Environmental impacts 12.8
Total 100.0
Figure 99: Criteria weights for base case
140
and non-potable water use. The TNPA Rig repair and Berth 205 scored fairly high as well (71),
as did Afrisam – Phase 1 (63). The lowest scoring projects overall were the Mineral beneficiation
plant (41) and ArcelorMittal (48). The Industrial Development Zone (IDZ) scored a relatively
average 57 points.
Some of the municipal representatives were at first surprised that Elandsfontein ended up being
the highest scoring project. Upon reflection, however, they realised that it made sense given the
chosen criteria, scores and weights. The project should use a very small amount of potable water
and a significant quantity of non-potable water, both positives. It should also employ significant
numbers of unskilled workers and its potential to benefit the local economy was thus deemed to
be relatively high. The environmental criterion was the only one where the project did not do well,
but this criterion was not given a prominent weighting in this particular decision-making setting.
This illustrates that from the context of the municipal officials evaluating these projects, that there
is an appetite to trade off environmental impacts in favour of improved socioeconomic outcomes
when limited water resources are required.
Another interesting result was the IDZ achieving an average weighted score relative to the other
alternatives, despite the project being a strategic priority for both provincial and national
government. However, again, this score was clearly explained through its relatively high potable
water use, the lack of requirements for non-potable water, and although there are expected to be
a high number of jobs generated, these were not seen to be highly compatible with the existing
skill-sets of the local population. Therefore many of these jobs will likely result in in-migration to
the area, increasing pressure on existing municipal services.
Figure 100: SBLM project alternative scoring results
It is important to note that these results are purely illustrative. The projects considered are
only a subset of the 31 projects included in the WCIP, and so the conclusions that can be
drawn from this ranking are limited. Furthermore, the tool has been designed to be used
within a broader decision-making process – one which preferably involves stakeholder
engagement and which occurs within the context of exploring options for supply-side
interventions, some of them involving public-private partnerships, which could meet the needs of
a growing industrial sector in SBLM.
Weighted Score
Base case
IDZ 57
Afrisam - Phase 1 63
Elandsfontein 1 81
ArcelorMittal 48
Mineral beneficiation plant 41
TNPA Rig repair at Berth 205 71
Alternative
141
7.5. Conclusions and insights
Achieving sustainable development in the context of limited resources is an immense challenge
facing municipalities. The implications are that trade-offs may need to be made. However the
consideration of trade-offs should be done in a systematic and transparent manner. The use of
MCDA to guide this decision-making process provides a vehicle for dialogue, cross-sectoral
collaboration and exposes the preferences of the municipal officials that may not otherwise be
apparent.
The MCDA process undertaken with the Saldanha Bay municipal officials was highly iterative
and collaborative in nature, with discussions occurring between departmental officials at all
stages of the process. It also evolved as the process went along, with criteria being added,
amended and removed. This is the flexibility that MCDA allows. Then when the results from the
WCIP sample set of projects were presented, they were not what was expected by the
participants in the process. These counter-intuitive results were examined carefully, but when
reviewed, it became apparent why the results were generated – and they were in complete
alignment with the preferences of the officials generating the results. That was the power of the
process – it makes the preferences of the decision-makers explicit.
However, there are also limitations to using MCDA. Generating the value functions and the
weightings can be quite complex if the criteria are not straight-forward. The weighting process in
particular was quite tricky, hence the ability for a “manual override” function. This could likely be
improved upon in later iterations of the tool. The use of Excel as the tool’s software platform
was deliberate as it is an easily accessible, it does not require any additional software licences,
nor is there a knowledge hurdle to using it. However there are limits to what Excel is able to
perform and the value functions, when exponential or logarithmic, were not performing
according to expectations. This meant that some values had to be manually generated. Utilising
an MCDA software may have solved some of these issues, but the project team was unable to
find an open-source or free MCDA software to use.
7.6. Way forward
The municipality has been provided with three Excel files:
1. A blank MCDA tool that could be used to generate results on any multi-factored decision.
This tool could be utilised in other municipalities for any decision that they wish the model
using MCDA. The tool has been developed with a key resource constraint around which
other criteria are weighted, but this resource (or reference criterion) could also easily be
cost.
2. The MCDA tool with the criteria and value functions specific to Saldanha Bay, but without
any project alternatives included. This tool version therefore allows the municipality to
quickly and easily evaluate project alternatives using the data available. The municipal
officials can either use the tool to compare competing development applications, or they
can start building a database of development applications, and for every new application
that is received, they can compare to previous applications in order to benchmark its
performance.
3. The MCDA tool with the WCIP projects included (as described in the section above) that
can be referred to in order to provide a benchmark for new projects, or it could be added
to as new project applications are submitted.
142
The municipality has also been provided with a detailed handover document to assist in the usage
of the files. Should copies of the files be required, please email [email protected]
In order to operationalise the use of the tool, support is being sought from the Saldanha Bay
councillors and the newly appointed Municipal Manager. It is important that political support for
this type of approach is secured, so that the officials involved in the decision-making process
are able to effectively allocate water to new developments in a manner consistent with the
development aspirations of the politicians.
143
8. Project insights and successes
8.1. Analytical insights
Currently the economic impact of water resource decisions, and the water resource implications
of economic decisions, are not fully considered. Water is taken into account in economic planning
(and vice versa), but generally as independent variables, rather than accommodating the co-
dependence of water and economics. Optimally integrating water resources and economic
development planning would see:
i) available resources allocated for optimal benefit, including consideration of economic
development and job creation;
ii) the economic benefit and water intensity of different water uses informing
development decisions;
iii) the ability of future developments to fund water resources interventions taken into
account in the feasibility of interventions; and
iv) overall water resources challenges taken into account in regional development
decisions.
What is needed to realise the type of integrated planning outlined above is largely provided for in
legislation. However, there are significant shortcomings in the processes through which the
legislation is implemented, resulting in them falling short of their intent and potential. Not all private
developments are known during the development of a municipal Integrated Development Plan
(IDP), and consequently considered during budgeting, which can lead to a situation where the
local municipality (LM) is incapable of providing sufficient water resources for development. More
fundamentally, in most cases the Water Services Development Plan (WSDP) is failing in its
consideration of water resources (i.e. supply options) and only considers service requirements,
essentially assuming an increase in the resource supplying that service is feasible. In the case of
the Berg Water Management Area (WMA), the Berg-Olifants Catchment Management Agency
(CMA) is not established, hence there is no Catchment Management Strategy (CMS) to support
the IDP and WSDP process. Contributions to the IDPs, including provincial oversight, are also
often rushed and serve to confirm a legislative requirement, but lack meaningful input from both
government (at local, provincial and national level) and the private sector.
Water allocation, which is the mandate of the national Department of Water and Sanitation (DWS)
in the absence of a CMA, functions as a licensing or authorisation process. If a water use licence
application meets the necessary application requirements, and there is water available to allocate,
the application is passed. The DWS has no scope (in its mandate) to, for example, wait for other
potentially more economically beneficial uses of the same (limited) water resource and allocate
water to this later application or benchmarks against which to compare relative merit of
applications.
On a local scale (and where water resources are local), it is the responsibility of a LM to ensure
that economic planning in the IDP is in line with water resources planning contained in the WSDP
(given the role of a Water Service Authority (WSA)). The reality, however, is that LMs are generally
not capacitated to plan accordingly. The WSAs are generally under-performing in terms of water
resources planning, and over-relying on DWS to provide this role, which it provides only for
regional schemes. Where adequate plans are made by WSAs, implementation slows due to
hurdles such as financing. In the Berg WMA, as with many other areas, water resources are
144
shared between many LMs, district municipalities (DMs) and a metropolitan municipality. Because
of the regional nature of the scheme, DWS does coordinate planning for water resources
interventions to reconcile future demand on the Western Cape Water Supply System (WCWSS).
However, this coordination is also insufficient i.e. not at the level required, given that competing
applications have been submitted from various WSAs that are using the WCWSS. The DWS has
been recommended to urgently resolve these competing applications, but not specifically the lack
of coordination which leads to such competition (DWS, 2015a).
The various assessments of the governance structures, done as part of this study, suggest that
provincial government is the most appropriate body to provide this strategic oversight and
integration role. Provincial government has the potential to identify and recognise threats to
development plans as a consequence of (water) resource constraints, and provide support to LMs
- both in terms of appealing for funding from national government for local water resource
development and by providing capacity and support in driving these processes (and in the
necessary water resources assessments). In the case of the Berg WMA, this oversight role should
not only focus on LMs requiring capacity support. The water resource needs and development
within City of Cape Town (CoCT) also needs to be considered by provincial government, in terms
of implications for the province. A more systemic/holistic view of the regional water needs and
potential sources would identify the potential for economies of scale, which are currently not being
adequately considered between WSAs within the Berg to prioritise the most appropriate water
resources infrastructure, prior to the related water use licence applications (WULAs) being
submitted to DWS for consideration.
In this study, an assessment of the value that water provides to the economy of the Berg WMA
was conducted and, as with other similar studies, the results showed that water utilised for
agricultural production produces less economic value than water used in urban sectors. This is
reflected for both employment and GVA. Furthermore, this value is not equally distributed across
the region, with the type of crops grown determining some of this difference, along with climatic
conditions. Drakenstein, Swartland and Stellenbosch generate the greatest value from their
agricultural water, while the City of Cape dominates in terms of value generated from urban water.
A structural analysis of the water intensity of an economy was also conducted in order to
understand how resilient the overall economy and those of different municipal economies are to
water shocks and constraints. This also revealed that local economies in the region have highly
varied water intensities. The combination of the value of water used and the intensity of this water
use is important to consider. A reduction in water supply to a largely agricultural economy will not
have the regional economic impacts that the same proportional reduction would have for a largely
tertiary economy, like the City of Cape Town, however, the local impacts would be felt severely.
(However, this does not consider the interdependence between rural and urban economies.) The
local impact can be seen in the current drought where the agricultural sector has been hit hard by
water restrictions and the lack of rainfall with job losses estimated to be as high as 50,000 in the
Western Cape (Evans, 2017). When considering the impact that climate change may have on the
water demands in the region, it was clear from the analysis that climate change will increase
agricultural water requirements. Together with growing urban water demands, future water
demand in the Berg WMA will exceed existing supply.
The findings from the project highlight that, without significant supply augmentation, the
development ambitions of the region will be severely constrained. There are augmentation
plans included in the WCWSS reconciliation strategy, however, with the exception of raising the
145
wall of the Voëlvlei Dam, all the schemes are intended for the City of Cape Town. This makes
sense when considering the impact that Cape Town has on the regional economy. But the
water intensity of the other local municipalities in the Berg WMA cannot be ignored when
considering how their local economies may deal with water scarcity. Furthermore, the current
(2015 - 2017) drought has highlighted that complete reliance on surface water does not provide
resilience in the midst of a changing climate - diversification of supply thus needs to be
prioritised. In this context, the analysis shows that the West Coast municipalities of Swartland,
Saldanha Bay and Bergrivier emerge as the municipalities where water is likely be to a
significant constraint to future development, unless water supply augmentation options are
urgently developed. Swartland and Saldanha Bay are rapidly growing areas, with significant
development proposed for Saldanha Bay, with focus on the Industrial Development Zone (IDZ).
However, these WSAs are already using more water than they have been allocated and have
highly water intense economies. Without a significant structural shift in the resource base
underlying their economy or an increase in their local water supply and/or active regional
optimisation/coordination to obtain solutions that benefit the region as a whole, these
economies will struggle to meet their development goals.
8.2. Project successes
Although the impact of the project has not been specifically assessed, there are several
qualitative indications that the project findings, and the project process itself, have had positive
impacts in forging greater integration between water resources development and economic
development planning in the Berg WMA. The project outputs have been incorporated in a
number of planning documents for the region, including the WCWSS reconciliation strategy, the
Municipal Economic Review and Outlook (MERO) and the Greater Saldanha Regional Spatial
Implementation Framework (GS RSIF). This implies that water resource planners and
development planners are paying closer attention to the potential impact that water has on
economic development and vice versa.
In recent engagements with stakeholders, there is an increasing acknowledgement that water
and economic development cannot be planned in isolation. During early engagements, not all
stakeholders saw the necessity for better integrated water and economic development planning.
Some thought some aspects being suggested by the project team were irrelevant, being
handled already, or too complex to meaningfully assess, and issues over mandate, turf and
necessity muddied debate. However, the conversations occurring in the latter part of the project
highlighted that the importance of integrated water and economics is now accepted. The project
(along with other initiatives and the 2015-2017 drought) has had a significant contribution to this
shift.
At a provincial government level, there is commitment from the Spatial Intelligence Unit of
Department of Environmental Affairs and Development Planning of the Provincial Government
of the Western Cape and the GIS unit at the Provincial Department of Agriculture (DoA) to
utilise the regional tool GIS files in their departments once the final report has been published.
This demonstrates that the research outputs have proven useful for their purposes, although
they would ideally like the analysis to extend to their entire province and not just the Berg WMA
(thereby also illustrating that they see value in this work).
The work conducted with Saldanha Bay Local Municipality (SBLM) on the development of the
tool, was also accompanied by other conversations and engagements relating to the local
146
municipality’s water resource options. The project team has spent a substantial time in the
Saldanha Bay area, working with civil society and business. This has allowed the project team
to build relationships in the area and gain a clearer understanding of the local challenges.
These relationships have been leveraged during the current drought, and the team has been
able to assist the municipality and businesses in the areas in their drought response strategies.
Furthermore, this study funded two Masters students. They not only delivered their research
dissertations, but also gained experience in applied research. Working with decision-makers on
a Masters project is difficult (see discussion below), and required multiple iterations of their
research proposals, methodology and analysis, but the students were able to deliver research
that proved to be useful and influential.
8.3. Methodology and process insights
One of the key innovations of the project structure was to provide an avenue for research
developed by academia to be guided and fully utilisable by government decision-makers.
GreenCape acted as the coordinating and facilitating entity for this project, attempting to bridge
the gap between the research requirements placed on the Masters students and the practical
insights required by government.
This was a challenging process at times:
The students needed to work with shifting research goals. They received feedback on their
research proposals, approaches and analysis from a number of different stakeholders, and
at times this was in conflict with their academic supervisor feedback. It was both confusing
and frustrating for the students to not have a clear research agenda and pathway.
The WRC requires that students contribute or are at least supported by the funding
allocated. Furthermore, the research funds available are generally inadequate to support
large teams or full-time researchers (in addition to students). The initial proposal for this
study therefore included time spent by students supported by their supervisors, with time
spent by GreenCape in coordination. The proposal team structure, however, had too much
reliance on the students to contribute to the project’s core research objectives, and the
supervisors were focussed on the individual dissertations rather than the overall project
output. A full-time researcher (per focus area) should have been included in the proposal to
deliver on the study, with supplementary research conducted by students.
The GreenCape project team, and the stakeholders that the team worked with, needed
research outputs that were useful and addressed their needs. However data constraints
were a concern and therefore the models that could be feasibly utilised were not necessarily
as academically robust as required for a Masters level degree (refer to the discussion in
chapter 5 about the marginal value of water versus the average value). This created tension
between developing tools that were useful versus academic outputs that meet university
requirements for methodology, originality and rigour, but that may never be used.
The challenge of combining academic requirements in an applied research environment was
overcome through splitting the reporting responsibilities of the students: they needed to deliver
on their academic requirements, while at the same time deliver on the project requirements. It
was not assumed that they would necessarily be the same outputs. GreenCape also played a
stronger role in coordinating and filtering stakeholder feedback, only reporting on aspects that
were relevant. GreenCape also took on a number of the research pieces in-house. So, although
the WRC intends to support and capacitate students with the research funds, this is not always
147
practical especially with a project of this nature that has direct and immediate relevance to and
input from its ultimate beneficiaries.
Utilising multiple criteria decision analysis (MCDA) to develop the SBLM water allocation
prioritisation tool allowed for greater flexibility and inclusion of different factors in comparison to
a cost-benefit analysis (CBA) as originally envisioned. It also demanded a cross-sectoral
engagement with a number of different municipal departments. The use of MCDA is also much
more transparent than CBA and this makes preferences explicit, opening up for further
dialogue. It also does not simply produce an “answer” or a figure, but rather requires
engagement in the process. However, this approach also came with its own complications as it
is theoretically more complex than a CBA, posing challenges for the project team, but the team
was fortunately assisted by an international MCDA expert (on a voluntary basis). The
development of the tool in Excel also required a number of workarounds that could have been
avoided had an MCDA software had been utilised. However, programming the tool in Excel
enabled the project team to gain in-depth understanding of the MCDA methodology, which had
the benefit that the team could also engage the stakeholders in a manner that allowed the tool
to be used correctly and not treated like a “black box” and thereby realise the wider benefits of
MCDA i.e. to enable understanding of preferences and their implications and to promote debate
among stakeholders.
8.4. Concluding Remarks
Overall the project is considered a success on the basis of having developed tools that meet the
needed of the intended beneficiaries, enabled capacity development for water sector challenges
and allowed knowledge sharing at a national level (primarily through conference presentations
and publications). Significantly, the project has had impact in terms of creating awareness of the
need for integrating water resource planning and economic development planning – there is
clear evidence of a change in awareness and mindset within key institutions and individuals
involved in water resource planning and economic development planning in the Berg WMA, and
agreement to adopt the tools in two provincial government departments. The work is considered
replicable in other water constrained catchments, and with both the key institutions (ACDI and
GreenCape) committed to further dissemination of the work and continued engagement with
relevant stakeholders in the Berg WMA, the impact of this project is expected to be extended
further in future.
148
9. Recommendations
9.1. Planning approaches
Based largely on the insights and way forward provided in section 4.5 and 4.6, the following
actions are recommended to achieve better integrated water resources and development
planning in the Berg WMA:
The WSAs in the Berg WMA (primarily municipalities) require support for water resources
planning, and in the absence of a regional water utility or water board (both of which
should be explored as possible avenues to resolve this issue), this support is best
provided at a provincial level by the WCG. However, tension remains with DWS regarding
the WCG’s involvement in water resources as this is not within their mandate. A facilitated
discussion between the WSAs in the region, WCG and DWS over mandate and the
potential for the formation of a separate entity (such as a water board) is recommended.
This process should ideally result in an agreement over an implementation protocol, as
per the Intergovernmental Relations Framework (RSA, 2005). Implementation protocols
are frameworks between organs of state designed to meet the challenge of joint work.
This framework is a tool or instrument for the practical application of co-operation and co-
ordination between organs of state, stakeholders or other agencies. An implementation
protocol should:
o Set out clear outcomes of the joint work of the three spheres of government
o Detail who is responsible for what task
o Determine what resources are required for the task at hand and who will provide
them
o Set performance indicators
o Put in place oversight mechanisms to ensure that outcomes are achieved.
An implementation protocol for a Berg WMA regional water utility could be coordinated
through the Western Cape Sustainable Water Management Plan (SWMP), and would
ideally involve the pooling of resources from local government, provincial government
and DWS for a feasibility study to assess the most appropriate water resource
development options (see point below).
Coordinated planning for future water resources interventions for those WSAs supplied
by the WCWSS must be strengthened. Competing applications are evidence that
coordination is lacking. There is currently a reliance on the WCWSS reconciliation
strategy, but local municipalities (including the City of Cape Town) will need to develop
their own local supplies due to a lack of viable surface water augmentation options. These
local schemes should be evaluated in comparison to the potential for regional schemes
that may allow for economies of scale. The most appropriate (cost effective and resource
efficient when considering the whole) water resources infrastructure should be
determined, prior to the related Water Use Licence Applications (WULAs) being
submitted to DWS for consideration. Achieving this may require a feasibility study to be
conducted (by the intergovernmental team noted above).
In order for WSAs to perform better in terms of water resources planning, water resources
planning must be strengthened in the Water Services Development Plan (WSDP)
process, and as such, water resources planning strengthened in the Integrated
Development Plan (IDP). Far greater emphasis needs to be placed on the availability of
water resources in the WSDP and IDP when a LM is considering its development plans.
This should be a requirement of provincial government when approving a local
149
municipality’s IDP, as should it be by DWS when approving the WSDP. It is
recommended that provincial government, (through the SWMP and other mechanisms),
oversee the water resources planned at LM level in WSDPs, coordinating these between
neighbouring LMs / DMs.
A gap (and a need) exists in the support to LMs in the preparation of project finance
applications to DWS (and other funding mechanisms, including commercial financiers).
The WCG should explore the options in terms of providing support to LMs for these
applications and/or the establishment of a regional water fund to assist in the financing of
water resource infrastructure.
Provincial and national development plans need to be more cognisant of the local
capabilities and water resource availability in the areas in which the development is
planned, crucially as it relates to the availability of water to support their development
ambitions. If sufficient water is not available, then a careful consideration should be made
about the suitability of a particular industry for the area (linked to its water intensity) versus
the cost/benefit of developing new water resources. To assist LMs to engage in debate
on such national (and provincial) plans, it is recommended that WCG use the outputs of
the regional tool to guide them in their regional planning and their assistance to the LMs,
and hence in terms of appropriate development trajectories to promote in their IDPs
Financing of off-budget water resource development continues to be a major hurdle when
the scheme is not being developed by DWS. Either Trans-Caledon Tunnel Authority
(TCTA) should be permitted to finance smaller projects that are not necessarily linked to
DWS (TCTA has been developing skills in desalination for the past few years and it may
prove capable of raising financing for these projects) or a public-private fund should be
created to assist local municipalities in their local water resource development plans.
There is a need for private sector investment in the water sector, and a regional water
utility could be a mechanism for this, while providing municipalities with access to
affordable finance.
9.2. Use of tools developed
The tools developed here can assist in achieving some of the planning related
recommendations made above. It is therefore recommended that:
WCG utilises the outcomes of the regional hydro-economic GIS model to promote particular
development trajectories at provincial scale, and to guide development choices at LM scale,
and to guide development of IDPs
LMs utilise the outcomes of the regional hydro-economic GIS model to understand the
impact of particular development choices, and to guide development of IDPs
SBLM utilise the developed MCDA tool to help inform discussions over permissions for
access to water
Other WMAs evaluate the potential to develop their own hydro-economic or MCDA tools
using the methodology described in this study. The blank MCDA tool can be provided in an
Excel format and used in a variety of situations to guide decision-making. The GIS hydro-
economic tool utilised publically available data, and although tricky to collate this data, it is
relatively straight-forward to re-run the model using the formulas described and a GIS
software (such as ArcGIS).
150
9.3. Further development of tools
Based on feedback gathered from stakeholders during the various stages of this project, it is
clear that there is demand for the tools developed, and that their outcomes will be implemented.
Furthermore, there have been requests made for certain add-ons or amendments to the tools
which are listed below.
that the analysis be expanded to a wider area, specifically the Western Cape as a whole
(i.e. beyond the Berg WMA)
that additional layers be added to the GIS tool that incorporates environmental and other
social criteria
that the current tool be updated when significant updates are made to the key input data,
including:
Inclusion of the new fly-over data when available (expected in 2018). In this regard, it
would be interesting to analyse changes in crop distribution and determine the difference
in water requirements, to assess whether farmers are responding to water scarcity in the
region.
Updates to water use data when validation and verification is completed expected in 2018
Update the model with the latest rainfall and temperature records for the drought period
(2015 – 2017) to see what the impact on the irrigation requirements may have been
9.4. Remaining gaps and suggestions for future research
The development of a hydro-economic tool as per the definitions included in section 5.1, Box 4
requires additional data that are currently not available in South Africa. A hydro-economic tool is
the best means of evaluating the socio-economic impacts of different water allocation scenarios.
In order to achieve this ultimate aim (which this project partially achieved), the following would
need to be considered:
Urban water usage classifications should be consistent between municipalities, and
ideally these classifications should be broken down into sectors as per the SIC codes
reported in their GVA and employment statistics. Multi-year data-sets would be required
to understand how changing water usage patterns result in different economic
outcomes.
In a similar manner, the agricultural census should be revived (the last census was in
2007) and regularly conducted. This is essential to understand the change in
agricultural revenue as per the change in water availability (among others).
Furthermore, all irrigated agricultural water use should be metered and made publically
available. This will allow for an accurate calculation of the marginal value of irrigated
agriculture and provide insight as to where the benefits from water can be best realised
in the agricultural sector. This data could also aid policy-makers looking to understand
irrigated water use in order to optimise it in constrained environments.
A key intervention that was not in the scope of the project, but kept on emerging in discussions,
relates to the financing of bulk water infrastructure and the leveraging of private sector funds.
This project highlights the urgency for a focused analysis on the potential for an innovative
financing solutions for the development of water resource infrastructure, which looks beyond
Public-Private-Partnerships (PPPs), such as a national water bond or bank. There was clear
appetite expressed by a number of stakeholders for this type of financing arrangement, but it is
unclear if analysis is currently being done on this topic.
151
10. References
Adnan, S. and Khan, A.H., 2009. Effective rainfall for irrigated agriculture plains of Pakistan.
Pak J Meteorol, 6(11), pp.61-72.
Arnold, J.G. and Fohrer, N., 2005. SWAT2000: current capabilities and research
opportunities in applied watershed modelling. Hydrological processes, 19(3), pp.563-572.
Ascough, G.W. & Kiker, G.A., (2002): “The effect of irrigation uniformity on irrigation water
requirements.” Water SA Vol. 28 No. 2.
Aspinall, R. and Pearson, D., 2000. Integrated geographical assessment of environmental
condition in water catchments: Linking landscape ecology, environmental modelling and
GIS. Journal of Environmental Management, 59(4), pp.299-319.
Bao, Q., Lin, P., Zhou, T., Liu, Y., Yu, Y., Wu, G., He, B., He, J., Li, L., Li, J. and Li, Y., 2013.
The flexible global ocean-atmosphere-land system model, spectral version 2: FGOALS-s2.
Advances in Atmospheric Sciences, 30, pp.561-576.
Bosman, D. (2016) of TCTA, personal communication with GreenCape. 7 June 2016
Bureau for Food and Agricultural Policy (BFAP). 2011. The contribution of the Agro-
Industrial Complex to Employment in South Africa. [online] Available at:
http://www.bfap.co.za/documents/research%20reports/BFAP%20-%202011%20-
%20The%20Contribution%20of%20the%20Agro-
Industrial%20Complex%20to%20Employment%20in%20South%20Africa.pdf
Chylek, P., Li, J., Dubey, M.K., Wang, M. and Lesins, G., 2011. Observed and model
simulated 20th century Arctic temperature variability: Canadian earth system model
CanESM2. Atmospheric Chemistry and Physics Discussions, 11(8), pp.22893-22907.
Claasen, J. 2015. Sharon fruit in SA – growing local and export sales, [online] Available at:
http://www.farmersweekly.co.za/crops/field-crops/sa-sharon-fruit/
Cole, MJ. 2017. Tracking inclusive sustainable development at multiple scales: South
Africa's safe and just operating spaces. D.Phil. dissertation. p332. University of Oxford.
Cole, M., Bailey, R., Cullis, J. & New, M. 2017. Spatial inequality in water access and water
use in South Africa. Accessible here:
https://www.dropbox.com/sh/79v2oqsr3eq61er/AADaQx81EB2CbCe0vENF5IuAa?dl=0
Conningarth Economists, 2012. Econometric model to predict the effect that various water
resource management scenarios would have on South Africa’s economic development.
Water Research Commission Report No. 1570/1/12, ISBN 978-1-4312-0255-3, June 2012
Crous, J.J. 2012. Managing olive yield and fruit quality under South African conditions,
Masters of Science in Agriculture: Stellenbosch
Department of Agriculture Forestry and Fishing, 2012. Statistics on Fresh Produce Markets
2011, [online] Available at: http://www.daff.gov.za/daffweb3/Portals/0/Statistics and
Economic Analysis/Statistical Information/Statistics on Fresh Produce Markets 2011 - MS
Excel.xls
152
Department of Economic Development & Tourism (DED&T), Provincial Government of the
Western Cape, 2016. West Coast Industrial Plan, Final Synthesis Report. January 2016.
Prepared by PDG, The Green House, and Strategies for Change
Department of Environmental Affairs and Development Planning (DEA&DP), Provincial
Government of the Western Cape, 2012. Western Cape Sustainable Water Management
Plan – 2012.
Department of Provincial and Local Government (DPLG), (2000). IDP Guide Pack
Department of Rural Development and Land Reform (DRDLR), (2012). Proposed Generic
Integrated Development Planning (IDP) Template
Department of Water Affairs (DWA) (2013a). National Water Resource Strategy 2nd Edition,
June 2013
Department of Water Affairs (DWA) (2013b). Strategic Plan: 2013/14 to 2017/18
Department of Water Affairs (DWA) (2014) Water Boards as Regional Water Utilities
(Institutional Arrangements to Facilitate Turnaround: Legal, Financial and Other Implications
Department of Water Affairs and Forestry (DWAF) (2010). Water for Growth and
Development, Version 7.
Department of Water Affairs and Forestry, South Africa. 2004. Berg Water Management
Area: Internal Strategic Perspective. Prepared by Ninham Shand (Pty) Ltd in association
with Jakoet and Associates, Umvoto Africa and Tlou and Matji, on behalf of the Directorate:
National Water Resource Planning. DWAF Report No P WMA19/000/00/0304
Department of Water Affairs and Forestry, South Africa. 2006. A Guide to Verifying the
Extent of Existing Lawful Water Use. Edition 2.1, November 2006
Department of Water Affairs and Forestry, South Africa. 2007. Watern Cape Water Supply
System: Reconciliation Strategy Study. Volume 1 of 7: Reconciliation Strategy, Final, June
2007. Prepared by Ninham Shand (Pty) Ltd in association with UWP Consulting, on behalf
of the Directorate: National Water Resource Planning. DWAF Report No P
WMA19/000/00/0507
Department of Water Affairs and Forestry, South Africa. 2008. The Assessment of Water
Availability in the Berg Catchment (WMA 19) by Means of Water Resource Related Models:
Report 4 (Land Use and Water Requirements): Volume 1 (Summary Report). Prepared by
Ninham Shand (Pty) Ltd in association with Umvoto Africa on behalf of the Directorate:
National Water Resource Planning. DWAF Report No P WMA19/000/00/0608
Department of Water and Sanitation (DWS) (2015b). Revision of the Pricing Strategy for
Water Use Charges in Terms of Section 56(1) of the National Water Act, 1998. Notice 1154
of 2015. Government Gazette No 39411
Department of Water and Sanitation (DWS) (2015c). Briefing the Portfolio Committee on
Water and Sanitation. Presented by: Margaret-Ann Diedricks, Director General. 03 June
2015. Available at: http://pmg- assets.s3-website- eu-west-
1.amazonaws.com/150603waterinvestment.pdf accessed 25 July 2017
Department of Water and Sanitation (DWS) South Africa, (2015a). Department of Water and
Sanitation, South Africa. 2015. Support to the Continuation of the Water Reconciliation
153
Strategy for the Western Cape Water Supply System: Status Report October 2015.
Prepared by Umvoto Africa (Pty) Ltd in association with WorleyParsons RSA on behalf of
the Directorate : National Water Resource Planning.
Department of Water and Sanitation (DWS), South Africa. 2017b. DEVELOPMENT OF THE
NATIONAL WATER & SANITATION MASTER PLAN Presentation at Berg-Olifants Regional
Work session Presented by: Patrick Mlilo Director: National Water Resource Planning. 27
JULY 2017
Department of Water and Sanitation (DWS), South Africa. February 2017a. Determination of
Water Resource Classes and Associated Resource Quality Objectives in the Berg
Catchment: Determining Environmental Water Requirements and changes in Non Water
Quality Ecosystem Goods, Services and Attributes. Project Number WP10987. DWS Report
NO:RDM/WMA9/00/CON/CLA/0217
Department of Water and Sanitation, South Africa. 2014. Implementation of a Water
Resource Classification System. DWS Brochure: Available at
http://www.dwa.gov.za/Documents/WRC/WRCSBrochure.pdf accessed 25 July 2017
Dollar E. S. J., Nicolson C. R., Brown C. A., Turpie J. K., Joubert A. R., Turton A. R., Grobler
D. F., Pienaar H. H., Ewart-Smith J. and Manyaka S. M. 2010. Development of the South
African Water Resource Classification System (WRCS): a tool towards the sustainable,
equitable and efficient use of water resources in a developing country. Water Policy, 12:
479–499.
DWA, 2009, Water For Growth and Development Framework, version 7
DWA, 2010 Integrated Water Resources Planning for South Africa, A Situation Analysis,
2010. JA van Rooyen and DB Versfeld. DWA report number P RSA 000/00/12910. 2010
Evans, J. 2017. Work drying up in Western Cape drought. (Online). Available:
https://www.news24.com/SouthAfrica/News/work-drying-up-in-western-cape-drought-
20171025
Geneletti, D. 2007. Multi Criteria Analysis. ValueES (Online). Available:
http://aboutvalues.net/data/method_navigator/values_method_profile_multi_criteria_analysis
.pdf [Accessed 08-11-2016].
GreenCape. (2015). Synthesis Report on Water as a Constraint on Economic Development,
2014-2015. Compiled by H. Seyler.
Haigh, E., H. Fox, H. Davies-Coleman and D. Hughes. 2008. The Role of Local Government
in Integrated Water Resources Management Linked to Water Services Delivery. WRC
REPORT NO 1688/1/08 ISBN 978-1-77005-820-0. August 2008
Harou, J. J., Pulido-Velazquez, M., Rosenberg, D. E. & Medellín-Azuara, J., 2009. Hydro-
economic models: Concepts, design, applications, and future prospects. Journal of
Hydrology, Volume 375, pp. 6247-643.
Hortgro 2012, Key Deciduous Fruit Statistics 2011, Paarl
Janse van Vuuren & Pineo, 2014. GreenCape 2014-2015 Synthesis Report on Regional
Resource Flow Model. Available from http://www.green-
cape.co.za/content/focusarea/resource-efficiency
154
Jensen, M.E., 1973. Consumptive use of water and irrigation water requirements. ASCE.
Jensen, M.E., Burman, R.D. and Allen, R.G., 1990. Evapotranspiration and irrigation water
requirements. ASCE.
Ji, D., Wang, L., Feng, J., Wu, Q., Cheng, H., Zhang, Q., Yang, J., Dong, W., Dai, Y., Gong,
D. and Zhang, R.H., 2014. Description and basic evaluation of BNU-ESM version 1. Geosci.
Model Dev. Discuss, 7(2), pp.1601-1647.
Kimaite, F. M., 2011. A Hydro-Economic Model for Water Resources Assessments with
Application to the Apalachicola-Chattahoochee-Flint River Basin, Doctor of Philosophy,
Georgia Institute of Technology: Georgia.
Kneale, P.E., 1991. Principles of Hydrology, RC Ward and M. Robinson, McGraw‐Hill, 1989
ISBN 0‐07‐707204‐9. Earth Surface Processes and Landforms, 16(3), pp.283-283.
Kotzé, M. 2012, Pomegranate Market Analysis. POMASA Information day [online] Available
at: www.sapomegranate.co.za/wp-content/uploads/2013/02/POMASA-newsletter-2-of-
2012.pdf POMASA 2012
KwaZulu-Natal Department of Agriculture and Rural Development, nd. Expected Yields,
[online]
Available at: http://www.kzndard.gov.za/images/Documents/Horticulture/Veg_prod/expected
_yields.pdf
Leenhardt, D., Trouvat, J.L., Gonzalès, G., Pérarnaud, V., Prats, S. and Bergez, J.E., 2004.
Estimating irrigation demand for water management on a regional scale: I. ADEAUMIS, a
simulation platform based on bio-decisional modelling and spatial information. Agricultural
Water Management, 68(3), pp.207-232.
Li, L., 2014. An overview of BCC climate system model development and application for
climate change studies. Journal of Meteorological Research, 28(1), pp.34-56.
Makhanya NO v Goede Wellington Boerdery (Pty) Ltd and Another (230/12, 233/12) [2012]
ZASCA 205; [2013] 1 All SA 526 (SCA) (30 November 2012)
Mekonnen, M. M. & Hoekstra, A. Y., (2011), The Green, Blue and Grey Water Footprint of
Crops and Devived Crop Products. "Hydrology and Earth System Sciences, 15 (5), p1577-
1600.
Monteith, J.L., 1965, July. Evaporation and environment. In Symp. Soc. Exp. Biol (Vol. 19,
No. 205-23, p. 4).
Mukadama 2014, SA steps up to the plate in olive oil industry, [online] Available at:
https://mg.co.za/article/2014-10-16-source-your-olive-oil-in-sa
Muller, J. (2017) Estimating the Marginal Value of Agricultural Irrigation Water, Masters of
Commerce, University of Cape Town, Cape Town
Muller, M. 2013. The Wicked Challenge of Sustaining South Africa’s Water Security. South
African Academy of Engineers (SAAE) Open Lecture, 7 November 2013.
National Planning Commission (NPC) (2011). National Development Plan Vision for 2030
Palmer, T. 2014. New paradigm for implementing the National Water Act. Key Note Lecture
at Catchment management: enhancing the understanding workshop 27th May 2014, WISA
155
Conference. And: Towards a New Paradigm in water resources management. Lecture at
Integrated Water Resources Management and RPMS workshop 27th May 2014 WISA
Conference
Pegasys. (2012). Review of Financing Mechanisms for infrastructure. Report submitted to
Department of Water Affairs as part of DWA project WP10465: ‘Project to Revise the Pricing
Strategy for Water Use Charges and Develop a Funding Model for Water Infrastructure
Development and Use and a Model for the Establishment of an Economic Regulator’. 29
June 2012.
Pegram G., and H. Baleta. Water in the Western Cape Economy. Progress report to the
Water Research Commission, WRC Project No K5-2075. Draft, December 2013.
Provincial Government of the Western Cape (WCG), 2013. Green is Smart: Western Cape
Green Economy Strategy Framework.
Provincial Government of the Western Cape; Department of Local Government (DLG),
(2000). IDP Guide Pack
Quantec. 2016a. Regional Output and GVA at basic prices by industry and 2011 municipal
level (ward-based region for metros) 1995-2015 (v2 26Sep16)
Quantec. 2016b. Employment and compensation by skill level, industry and 2011 municipal
level (ward-based region for metros) 1993-2015 (v2 26Sep16)
Republic of South Africa (RSA). (1956). Water Act (Act No. 54 of 1956).
Republic of South Africa (RSA). (1996). Constitution of the Republic of South Africa (Act No.
108 of 1996).
Republic of South Africa (RSA). (1997). Water Services Act (Act No. 108 of 1997).
Government Gazette, (18522).
Republic of South Africa (RSA). (1998). National Water Act (Act No. 36 of 1998).
Government Gazette, (19182).
Republic of South Africa (RSA). (2000). Municipal Systems Act (Act No. 32 of 2000).
Government Gazette, (21776).
Republic of South Africa (RSA). (2003). Local Government: Municipal Structures
Amendment Act (Act No. 1 of 2003). Government Gazette, (24730).
Republic of South Africa (RSA). (2005) Intergovernmental Relations Framework Act, 2005
(Act No. 13 of 2005)
Republic of South Africa (RSA). (2013). Spatial Planning and Land Use Management Act 16
of 2013
Republic of South Africa (RSA). 2014. Medium-Term Strategic Framework (MTSF) 2014 -
2019. Available at: http://www.gov.za/sites/www.gov.za/files/MTSF_2014-2019.pdf
Riemann, K., D. Louw, N. Chimboza and M. Fubesi, 2011. Groundwater Management
Framework. Water research commission report no. 1917/1/10, February 2011.
Schulze, R.E., Maharaj, M. and Moult, N. 2007. Reference Crop Evaporation by the
Penman- Monteith Method. In: Schulze, R.E. (Ed). 2007. South African Atlas of Climatology
156
and Agrohydrology. Water Research Commission, Pretoria, RSA, WRC Report 1489/1/06,
Section 13.3.
Serra, P., Salvati, L., Queralt, E., Pin, C., Gonzalez, O. and Pons, X., 2016. Estimating
Water Consumption and Irrigation Requirements in a Long‐Established Mediterranean Rural
Community by Remote Sensing and Field Data. Irrigation and Drainage, 65(5), pp.578-588.
South African Table Grape Industry, 2011. SATI Statistical Booklet 2011, Paarl: SATI
South African Wine Industry Information and Systems (SAWIS). 2015. Macro-economic
Impact of the Wine Industry on the South African Economy, [online] Available at:
http://www.sawis.co.za/info/download/Macro-economic_impact_study_- Final_Report_Versi
on 4 30Jan2015.pdf
Stamm, G.G., (1967) “Problems and procedures in determining water supply requirements
for irrigation projects. Chapter 40 in Irrigation of Agricultural Lands by Hagan et al.
Wisconsin, American Society of Agronomy.
Statistics South Africa & DEADP 2017b. 2016 Community Survey. Draft Data. Population
and Households
Statistics South Africa 2010, Census of commercial agriculture 2007, [online] Available at:
http://www.statssa.gov.za/publications/Report-11-02-01/Report-11-02-012007.pdf
Statistics South Africa 2017a. Quarterly Labour Force Survey – QLFS Q1:2017. [Online]
Available at: http://www.statssa.gov.za/publications/P0211/P02111stQuarter2017.pdf
[Accessed 13 September 2017].
Taylor, K.E., Stouffer, R.J. and Meehl, G.A., 2012. An overview of CMIP5 and the
experiment design. Bulletin of the American Meteorological Society, 93(4), p.485.
Thomas, A., 2008. Agricultural irrigation demand under present and future climate scenarios
in China. Global and Planetary Change, 60(3), pp.306-326.
Thompson, I., Riemann, K., and Hay, R. 2015. All Towns Reconciliation Strategy in the
Western and Eastern Cape – Achievements and Challenges. Paper presented at IMESA
Conference 27-30 October 2015, GrandWest, Cape Town. 10 pages.
Thornthwaite, C.W., 1948. An approach toward a rational classification of climate.
Geographical review, 38(1), pp.55-94.
Voldoire, A., Sanchez-Gomez, E., y Mélia, D.S., Decharme, B., Cassou, C., Sénési, S.,
Valcke, S., Beau, I., Alias, A., Chevallier, M. and Déqué, M., 2013. The CNRM-CM5. 1
global climate model: description and basic evaluation. Climate Dynamics, 40(9-10),
pp.2091-2121.
Watanabe, M., Suzuki, T., O'ishi, R., Komuro, Y., Watanabe, S., Emori, S., Takemura, T.,
Chikira, M., Ogura, T., Sekiguchi, M. and Takata, K., 2010. Improved climate simulation by
MIROC5: mean states, variability, and climate sensitivity. Journal of Climate, 23(23),
pp.6312-6335.
Watanabe, S., Hajima, T., Sudo, K., Nagashima, T., Takcmura, T., Okajima, H., Nozawa, T.,
Kawase, H., Abe, M., Yokohata, T. and Ise, T., 2011. MIROC-ESM 2010: model description
and basic results of CMIP 5-20 c 3 m experiments. Geoscientific Model Development, 4(4),
pp.845-872.
157
Weslander (2013). 14 February 2013 edition, page 11 [online] Available at:
https://issuu.com/weslander/docs/weslander_14-02-2013/11
West Coast District Municipality, 2010. Investigation into alternative water sources for the
West Coast District Municipality. Water Study Report, Final. Completed by Element
Engineers, October 2010
Western Cape Department of Agriculture (WCDoA), 2013. Winter Flyover Shapefile
Western Cape Government (WCG) (2014). Western Cape Provincial Spatial Development
Framework, March 2014
Wilby, R.L. and Wigley, T.M.L., 1997. Downscaling general circulation model output: a
review of methods and limitations. Progress in physical geography, 21(4), pp.530-548.
WRC (2016). Upgrading of SAPWAT3 as a Management Tool to Estimate the Irrigation
Water Use of Crops (WRC Project No K8/1115//4.)
WWAP (United Nations World Water Assessment Programme). 2016. The United Nations.
World Water Development Report 2016: Water and Jobs. Paris, UNESCO.
Yukimoto, S., Adachi, Y., Hosaka, M., Sakami, T., Yoshimura, H., Hirabara, M., Tanaka,
T.Y., Shindo, E., Tsujino, H., Deushi, M. and Mizuta, R., 2012. A new global climate model
of the Meteorological Research Institute: MRI-CGCM3—model description and basic
performance— Meteorological journal. Part 2, 90(0), pp.23-64.